Two distinct SWI/SNF complexes direct chromatin-linked transcriptional programs in Toxoplasma Authors: Dominic Schwarz, Benicio Tapia, Sebastian Lourido Published: 2026 ABSTRACT Chromatin remodeling complexes dynamically modify DNA accessibility to regulate gene expression during eukaryotic cell cycle progression, developmental transitions, and environmental adaptation. Higher eukaryotes have multiple remodeler subtypes; however, the coordination and functional specificity of these diverse complexes is not well understood. Apicomplexan parasites such as Toxoplasma gondii have a limited set of chromatin remodelers offering a divergent setting in which to explore the function of homologous complexes. These parasites have retained the Myb domain–containing proteins with homology to chromatin-associated regulators like SNF2h and SWI3. Here, a comprehensive analysis of the Myb protein family in Toxoplasma defines the composition of two SWI3 complexes with mutually exclusive ATPase homologs—TgSNF2a and TgSNF2b. Integrating transcriptomics with a custom chromatin-profiling strategy, we show that TgSNF2a is essential for the timely transcription of genes, while TgSNF2b ensures global transcriptional competency throughout the cell cycle and developmental transitions. Cell cycle–resolved chromatin profiling conclusively shows the shift from TgSNF2b to TgSNF2a occupancy when regulated genes transition from being poised to being actively transcribed. Our findings demonstrate that TgSNF2a and TgSNF2b perform distinct yet interdependent regulatory roles shaped by their chromatin context. This work offers new insight into the functional diversification of SWI/SNF complexes across eukaryotes. Introduction Apicomplexan parasites rely on intricate gene-expression programs, orchestrated by DNA-binding proteins and chromatin regulators, to colonize hosts and evade immune responses1–8. In Toxoplasma gondii, a model apicomplexan parasite, broad transcriptional changes shape key developmental transitions. During the lytic cycle, Toxoplasma tachyzoites undergo a tightly regulated cell cycle consisting of G1a, G1b and overlapping S, M, and C phases. G1a and G1b constitute roughly half of the cell cycle, dedicated to cellular growth and preparation for DNA replication. S phase initiates chromosome duplication, organelle biogenesis, and centriole replication, while the formation of daughter cells begins late in S phase and proceeds through closed mitosis. These sequential events are driven by transcriptional programs, with successive waves of gene expression ensuring the timely production of factors essential for daughter cell formation9–11. In addition to cell cycle–linked changes, large-scale transcriptional shifts define developmental stage transitions as parasites differentiate to chronic bradyzoites, sexual stages, or environmentally resilient oocysts necessary for transmission and recombination12. Together, these dynamic expression programs enable Toxoplasma to balance rapid proliferation with long-term persistence across diverse host environments. DNA-binding transcription factors have been extensively studied as key regulators of gene expression. In particular, AP2 (Apetala2/ethylene response factor) transcription factors have received considerable attention due to their lineage-specific expansion in apicomplexans and their critical roles in controlling developmental transitions and cell cycle progression13. However, chromatin structure constrains access to regulatory DNA elements and is shaped by nucleosome positioning, histone modifications, and higher-order chromatin organization14. Chromatin transitions frequently precede transcriptional changes, with certain regulatory boundaries becoming detectable at the chromatin level before transcription is initiated6. Both conserved and parasite-specific histone modifications have been reported in apicomplexan parasites, which are tightly regulated by histone-modifying enzymes7,15,16. Yet, the functional landscape of chromatin readers—proteins that recognize and interpret histone modifications to mediate downstream regulatory processes—remains largely unexplored in apicomplexans, as current models of chromatin regulation primarily emphasize the recruitment of remodeling complexes through interactions with DNA-binding AP2 transcription factors7,8,17. The mechanisms by which these factors access regulatory DNA regions remain poorly understood, highlighting a major gap in our understanding of gene regulation in these parasites. Our recent work pointed to Myb domain-containing proteins as compelling candidates for alternative chromatin-regulatory pathways in T. gondii18. Although some Myb proteins, such as Toxoplasma BFD1, Cryptosporidium Myb-M, and Plasmodium Myb1, retain classical DNA-binding functions, apicomplexans have predominantly retained Myb proteins of the SANT type, which lack DNA-binding capacity due to their negatively charged electrostatic surface3,18–20. SANT domains are commonly found in chromatin remodeling complexes, where they mediate interactions with histone tails and regulatory cofactors18. This suggests that apicomplexans may rely more heavily on chromatin-based regulatory mechanisms, with SANT-type Mybs potentially playing a central role in chromatin remodeling and gene regulation. In the present study, we comprehensively analyze the Myb domain-containing proteins of T. gondii and identify nine as essential with diverse phenotypes and transcriptional signatures. Among those, the SWI3 ortholog emerged as a central regulator that assembles into two distinct chromatin-associated complexes with broad genomic distribution and profound impacts on gene expression. Both complexes are distinguished by the incorporation of either TgSNF2a or TgSNF2b, SWI/SNF helicases that confer catalytic activity and establish regulatory potential. In canonical eukaryotic systems, SWI/SNF complexes are evolutionary conserved remodelers that regulate chromatin accessibility by coupling ATP hydrolysis to nucleosome repositioning or eviction, thereby modulating gene expression programs that control fundamental processes such as differentiation, proliferation, and DNA repair21,22. In Toxoplasma, these remodelers similarly contribute to key biological transitions with TgSNF2a- and TgSNF2b-containing complexes regulating cell cycle progression and developmental stage transitions. However, they diverge in sequence similarity and complex composition and—unlike the partially compensatory roles of the mammalian orthologs23—also display distinct regulatory activities. Results A diverse array of Myb domain-containing proteins is essential for the Toxoplasma lytic cycle Toxoplasma gondii encodes 15 Myb domain–containing proteins, of which five contain canonical DNA-binding Myb-type domains, and ten contain SANT-type domains typically associated with chromatin-related functions18. While Myb-type proteins include the master regulator of chronic differentiation BFD1, the SANT-type group contains orthologs of chromatin-associated factors such as Ada2 (Ada2a: TGME49_217050, Ada2b: TGME49_262420), SWI3 (TGME49_286920), RCOR1 (TGME49_321450), DMAP1 (TGME49_213890) or the recently characterized SNF2h (TGME49_321440) (Fig. 1A)3,8,18. To systematically characterize their role during the Toxoplasma lytic cycle, we utilized a high-throughput tagging strategy, paired with a minimal auxin-inducible degron (mAID) to profile subcellular localization and growth defects following auxin-induced conditional knockdowns (cKD; Fig. 1B)24. BFD1 was excluded as it has been extensively characterized3.Fig. 1A high-throughput tagging strategy reveals functional diversity of Toxoplasma Myb domain–containing proteins.A Phylogeny of Toxoplasma Myb domain–containing proteins18; DNA-binding Myb-type proteins (green squares) and SANT-type Myb proteins (purple squares). Numerals denote the number of Myb domain repeats. The right square indicates the phenotype score in a genome-wide CRISPR screen, reflecting the relative fitness contribution of each gene25. Bootstrap values of 1000 replicates are represented by circles. B Generation of conditional knockdown strains with high-throughput tagging with mNG-mAID-Ty, which targets proteins for degradation upon auxin treatment24. The pyrimethamine-resistance cassette (DHFR) enables mutant selection. C, D Plaque assays and vacuoles after 24 and 48 h of knockdown of Myb (C) or SANT (D) proteins, representative of at least two independent experiments. Immunofluorescence with antibodies against GAP45 (magenta), mNG or ALFA-tag (green), and Hoechst for DNA (blue). Asterisk denotes a strain tagged with 2×ALFA due to insufficient signal from mNG constructs. E Competition assays of fitness-conferring Myb (left) or SANT (right) proteins comparing the relative growth of mAID-tagged strains against the parental expressing IMC1-tdTomato24 in auxin-containing media. At the indicated time points, the proportion of fluorescent parasites within each competing population was measured by flow cytometry and normalized to the vehicle control. F Images of cell cycle arrest following CDC5L knockdown, representative of at least two independent experiments. Parasites were fixed and stained with antibodies against Centrin (yellow), CDPK1 (magenta), and mNG (green), and DNA visualized with Hoechst (blue). G Quantification of centrosomes per parasite nucleus of CDC5L knockdown parasites, quantified based on immunofluorescence analysis (F). Data represent the mean from two independent experiments. H Images of the TFIIIB differentiation phenotype following knockdown, representative of at least two independent experiments. Parasites were fixed and stained with antibodies against CDPK1 (magenta), and mNG (green), DNA was visualized with Hoechst (blue); the cyst wall was stained with DBL (yellow). I Quantification of DBL-positive vacuoles of TFIIIB knockdown parasites. Vacuoles were quantified based on immunofluorescence analysis (H). Data represent the mean from two independent experiments. A high-throughput tagging strategy reveals functional diversity of Toxoplasma Myb domain–containing proteins. A Phylogeny of Toxoplasma Myb domain–containing proteins18; DNA-binding Myb-type proteins (green squares) and SANT-type Myb proteins (purple squares). Numerals denote the number of Myb domain repeats. The right square indicates the phenotype score in a genome-wide CRISPR screen, reflecting the relative fitness contribution of each gene25. Bootstrap values of 1000 replicates are represented by circles. B Generation of conditional knockdown strains with high-throughput tagging with mNG-mAID-Ty, which targets proteins for degradation upon auxin treatment24. The pyrimethamine-resistance cassette (DHFR) enables mutant selection. C, D Plaque assays and vacuoles after 24 and 48 h of knockdown of Myb (C) or SANT (D) proteins, representative of at least two independent experiments. Immunofluorescence with antibodies against GAP45 (magenta), mNG or ALFA-tag (green), and Hoechst for DNA (blue). Asterisk denotes a strain tagged with 2×ALFA due to insufficient signal from mNG constructs. E Competition assays of fitness-conferring Myb (left) or SANT (right) proteins comparing the relative growth of mAID-tagged strains against the parental expressing IMC1-tdTomato24 in auxin-containing media. At the indicated time points, the proportion of fluorescent parasites within each competing population was measured by flow cytometry and normalized to the vehicle control. F Images of cell cycle arrest following CDC5L knockdown, representative of at least two independent experiments. Parasites were fixed and stained with antibodies against Centrin (yellow), CDPK1 (magenta), and mNG (green), and DNA visualized with Hoechst (blue). G Quantification of centrosomes per parasite nucleus of CDC5L knockdown parasites, quantified based on immunofluorescence analysis (F). Data represent the mean from two independent experiments. H Images of the TFIIIB differentiation phenotype following knockdown, representative of at least two independent experiments. Parasites were fixed and stained with antibodies against CDPK1 (magenta), and mNG (green), DNA was visualized with Hoechst (blue); the cyst wall was stained with DBL (yellow). I Quantification of DBL-positive vacuoles of TFIIIB knockdown parasites. Vacuoles were quantified based on immunofluorescence analysis (H). Data represent the mean from two independent experiments. We examined all Myb domain–containing proteins that were found to be fitness conferring in a genome-wide CRISPR screen25. Out of the eleven fitness-conferring Myb proteins, nine displayed clear growth defects by plaque assay, suggesting essential roles during the lytic cycle (Fig. 1C, D). Notably, TGME49_203380, a homolog of the co-chaperone DNAJC, did not localize to the nucleus, while TGME49_259860, a homolog of Telomeric Repeat-binding Factor 2 (TRF2), exhibited a distinct sub-nuclear localization—patterns that are consistent with their predicted functional homology (Fig. 1D)18,26,27. The absence of a plaquing defect for TGME49_203380 and TGME49_259860 indicates that these proteins may confer only a subtle fitness defect, detectable primarily under competitive growth conditions. To further assess the impact of Myb protein depletion on parasite fitness, we performed competition assays by co-culturing each cKD line with the parental strain expressing IMC1-tdTomato24 in the presence of auxin. The resulting population dynamics closely mirrored the severity of the phenotypes observed by microscopy. Mutants exhibiting pronounced morphological defects or cell cycle arrest dropped out rapidly, whereas those with milder or delayed phenotypes displayed more gradual declines (Fig. 1E; Source Data). Knockdown of nuclear and essential Myb proteins caused aberrant cell morphologies with two exceptions. Depletion of TGME49_275480—a homolog of the spliceosome component and cell cycle regulator CDC5L—resulted in a G1-phase arrest (Fig. 1F, G; Source Data)28–30. By contrast, depletion of TGME49_203950—a homolog of Transcription factor TFIIIB component B (TFIIIB) and Plasmodium spp. High Mobility Group Protein B3 (HMGB3)31 led to the formation of tissue cyst-like structures under standard growth conditions (Fig. 1H, I; Source Data). The Myb proteins lacking fitness defects were assessed for their involvement in stress-induced differentiation. Quantifying cyst wall staining with Dolichos biflorus lectin (DBL), demonstrated that the three dispensable Myb proteins, along with TGME49_203380 and TGME49_259860—which exhibited no defects in plaque formation—showed no defects in differentiation (Supplementary Fig. 1A, B; Source Data). INO80 and ISWI chromatin-regulator components share cell cycle signatures distinct from SWI/SNF To place the Myb proteins into the hierarchy of a transcriptional network, we performed transcriptomic profiling following knockdown for 6 or 12 h, focusing on those that were nuclear-localized and fitness-conferring. Principal component analysis (PCA) showed that a subset of knockdown samples separated clearly from the control, reflecting substantial transcriptional changes upon Myb protein depletion, while other knockdown samples displayed a more subtle divergence (Fig. 2A, Supplementary Fig. 2A–C, and Supplementary Data 1). For example, the knockdown of Myb proteins with homology to transcription factors, such as SNPC4 and TFIIIB, caused only modest transcriptional changes. By contrast, depletion of Myb proteins related to chromatin regulation and the highly conserved spliceosome-associated factor CDC5L resulted in widespread transcriptional dysregulation (Fig. 2A–C)29.Fig. 2Myb proteins associated with chromatin remodeling differentially impact the Toxoplasma cell cycle.A PCA was performed on the 500 most variable genes across all samples, based on DESeq278 variance-stabilized expression values from biological triplicates for each condition and time point. The first three principal components (PC) are shown. B Conservation analysis of full-length protein sequences for Myb proteins. For each protein and species, the lowest e-value hit was reported to indicate sequence conservation; hits exceeding the 0.01 threshold were visually marked to reflect a lack of significant homology between homologs. C Bar plot displaying numbers of significantly upregulated and downregulated genes following knockdown of individual Myb proteins, as determined by DESeq278. Differentially expressed genes were defined by an adjusted p value below 0.005 and an absolute fold change >2. The proportion of genes with low read counts are shown in light shading, and Myb proteins with predicted roles in chromatin regulation are annotated. D Schematic of the domain organization for three Myb proteins associated with ATP-dependent chromatin remodeling. Domains annotated according to the SMART database86: SANT domains, helicase domains (DEXDc & HELICc), and coiled coil (CC) regions. E Transcriptomic impact of ATP-dependent chromatin remodeler components on cell cycle-regulated gene expression. Mean log₂ fold change in RNA abundance of gene sets associated with distinct cell cycle phases significantly affected following 6 h of knockdown. Cell cycle gene sets based on ref. 33; set sizes indicated. F Pairwise comparisons of transcriptomic changes following 6 hr knockdown of proteins belonging to three distinct ATP-dependent chromatin remodeling complexes. Scatter plots display log₂ fold changes from RNA-seq data of genes with a Benjamini–Hochberg adjusted p value <0.005 by Wald test. Genes are color-coded according to their cell cycle-regulated classification as shown in (E). Dot size corresponds to the mean RNA-seq counts in vehicle-treated samples of the corresponding strains. G Immunofluorescence image of TgSWI3 after 48 h of knockdown, representative of two independent experiments. Parasites were stained with antibodies against GAP45 (magenta), mNG (green), and IMC1 (yellow) as well as Hoechst for DNA (blue). Myb proteins associated with chromatin remodeling differentially impact the Toxoplasma cell cycle. A PCA was performed on the 500 most variable genes across all samples, based on DESeq278 variance-stabilized expression values from biological triplicates for each condition and time point. The first three principal components (PC) are shown. B Conservation analysis of full-length protein sequences for Myb proteins. For each protein and species, the lowest e-value hit was reported to indicate sequence conservation; hits exceeding the 0.01 threshold were visually marked to reflect a lack of significant homology between homologs. C Bar plot displaying numbers of significantly upregulated and downregulated genes following knockdown of individual Myb proteins, as determined by DESeq278. Differentially expressed genes were defined by an adjusted p value below 0.005 and an absolute fold change >2. The proportion of genes with low read counts are shown in light shading, and Myb proteins with predicted roles in chromatin regulation are annotated. D Schematic of the domain organization for three Myb proteins associated with ATP-dependent chromatin remodeling. Domains annotated according to the SMART database86: SANT domains, helicase domains (DEXDc & HELICc), and coiled coil (CC) regions. E Transcriptomic impact of ATP-dependent chromatin remodeler components on cell cycle-regulated gene expression. Mean log₂ fold change in RNA abundance of gene sets associated with distinct cell cycle phases significantly affected following 6 h of knockdown. Cell cycle gene sets based on ref. 33; set sizes indicated. F Pairwise comparisons of transcriptomic changes following 6 hr knockdown of proteins belonging to three distinct ATP-dependent chromatin remodeling complexes. Scatter plots display log₂ fold changes from RNA-seq data of genes with a Benjamini–Hochberg adjusted p value <0.005 by Wald test. Genes are color-coded according to their cell cycle-regulated classification as shown in (E). Dot size corresponds to the mean RNA-seq counts in vehicle-treated samples of the corresponding strains. G Immunofluorescence image of TgSWI3 after 48 h of knockdown, representative of two independent experiments. Parasites were stained with antibodies against GAP45 (magenta), mNG (green), and IMC1 (yellow) as well as Hoechst for DNA (blue). We next compared Myb proteins with homology to chromatin-regulatory components. Comparative analysis showed SNF2h is well conserved among apicomplexans, yeast, fruit fly, and humans, while Ada2a shows moderate conservation, and DMAP1 and SWI3 are more divergent, with limited sequence similarity (Fig. 2B). PCA analysis showed that knockdown of TgSNF2h and TgDMAP1 induced comparable transcriptional responses, whereas depletion of TgSWI3 led to a distinct expression profile (Fig. 2A). TgSNF2h or TgDMAP1 6 hr knockdown resulted in a greater proportion of upregulated genes (TgSNF2h: 503; TgDMAP1: 535) compared to downregulated genes (TgSNF2h: 223; TgDMAP1: 310), whereas the depletion of TgSWI3 predominantly resulted in downregulation (126 genes) instead of upregulation (64 genes), underscoring differences in their regulatory impact (Fig. 2C). Structural features further indicate functional specialization. TgSNF2h contains the conserved ATPase domain characteristic of the catalytic core of the ISWI complex8. By contrast, TgDMAP1 and TgSWI3 lack ATPase domains and are presumed to function as cofactors within the INO80 and SWI/SNF complexes, respectively (Fig. 2D)32. Despite their shared involvement in chromatin remodeling, the transcriptional signatures of TgSNF2h, TgDMAP1, and TgSWI3 diverged substantially, particularly on genes expressed during specific stages of the parasite’s cell cycle, as defined according to ref. 33. Knockdown of TgSNF2h or TgDMAP1 primarily caused upregulation of S/M-specific transcripts and downregulation of G1a-specific transcripts, while TgSWI3 depletion led to global downregulation of cell cycle-specific transcripts (Fig. 2E, F). Consistent with their transcriptional impacts, depletion of TgDMAP1 and TgSNF2h resulted in similar phenotypes, characterized by cell cycle arrest and disrupted nuclear morphology (Fig. 1D). By contrast, parasites depleted of TgSWI3 retained the ability to undergo nuclear division but failed to complete cytokinesis (Fig. 2G). To investigate whether TgSNF2h, TgDMAP1, or TgSWI3 functionally cooperate with known transcriptional repressors, we explored their relationship with the TgMORC complex, a well-established repressive chromatin modifier directed by AP2 transcription factors7. Reanalysis of published TgMORC ChIP-seq data revealed distinct correlations between TgMORC occupancy at promoter regions and the transcriptional impact of different chromatin regulators. Specifically, genes upregulated upon TgDMAP1 knockdown exhibited higher TgMORC enrichment at promoter regions. By contrast, TgSNF2h displayed the opposite trend: downregulated genes were associated with elevated TgMORC occupancy, while upregulated transcripts corresponded to loci with reduced TgMORC binding (Supplementary Fig. 3A–D and Supplementary Data 2). Together, these findings point to a functional divergence among ATP-dependent chromatin remodelers in T. gondii. While ISWI- and INO80-associated factors (TgSNF2h and TgDMAP1) display opposing dependencies on the repressive TgMORC complex, TgSWI3 exhibits a transcriptional profile distinct from both and consequently a more complex relationship with TgMORC occupancy. TgSWI3 defines two distinct SWI/SNF complexes distinguished by two different ATPases SWI3 acts as a central scaffold in canonical SWI/SNF chromatin remodeling complexes, which regulate DNA accessibility for transcription, replication, and DNA repair. These remodelers function through the coordinated action of distinct modules—including ATPase and substrate recognition domains, that couple ATP hydrolysis to nucleosome repositioning or eviction21,34. To understand how SWI/SNF chromatin regulators operate in apicomplexan parasites, we set out to define the composition of the Toxoplasma SWI/SNF complex by co-immunoprecipitation mass spectrometry (co-IP MS) of TgSWI3 endogenously tagged with mNG. Comparison of pulldowns with an untagged control revealed a network of TgSWI3-associated proteins, including several factors homologous to conserved components of the SWI/SNF complex such as SS18 (TGME49_226900), SMARCD1 (TGME49_248120), BRD4 (TGME49_306460), ALP2a (TGME49_258050), SMARCB1 (TGME49_226350), and ARID (TGME49_246170). Notably, the co-IP included two DEAD-box helicases: TGME49_278440 and TGME49_320300. Reciprocal IPs with either helicase confirmed their interaction with TgSWI3 and several other SWI/SNF complex homologs (Fig. 3A and Supplementary Data 3).Fig. 3Two SWI/SNF chromatin remodeling complexes are defined by their different helicases, TgSNF2a or TgSNF2b.A Co-IP MS of endogenously TgSWI3-mNG, TgSNF2a-2×ALFA, and TgSNF2b-2×ALFA. Plotted abundances of proteins identified with >2 unique peptides for n = 2 or 3 biological replicates for TgSWI3 or TgSNF2a/b, respectively. Significantly enriched proteins are colored according to their Benjamini–Hochberg corrected two-way ANOVA p-values from abundance ratio analysis. B Heatmap of proteins co-purified with TgSWI3, TgSNF2a, or TgSNF2b in at least two pulldowns, based on analysis in (A). Tiles colored by their adjusted p-value in blue or their conservation in representative genomes, the latter reports the lowest e-value found. C Schematic of domain architectures from InterProScan37. HsBRG1 and HsBRM are shown for reference, alongside AlphaFold338 predictions of the DEXDc and HELICc domains from TgSNF2a (orange) and TgSNF2b (blue); overall RMSD of 5.45 Å and domain-specific RMSDs of 2.42 Å or 0.54 Å for DEXDc or HELICc, respectively. D Immunoblot analysis of stability during reciprocal knockdown of TgSNF2a-mAID or TgSNF2b-mAID for increasing lengths of time, representative of two independent experiments. Blots probed with anti-ALFA to monitor the abundance of the non-degraded helicase, anti-V5 to confirm the knockdown, and anti-SAG1 as a loading control. E Competition assays of TgSWI3, TgSNF2a, and TgSNF2b mAID-tagged strains comparing the relative growth against the parental strain expressing IMC1-tdTomato24 in auxin-containing media. At each time point, the proportion of fluorescent parasites was normalized to the vehicle control. F Immunofluorescence images of TgSNF2a and TgSNF2b knockdown parasites following 72 h of treatment, representative of at least two independent experiments. Fixed parasites were stained for GAP45 (magenta), IMC1 (yellow), mNG (green), and Hoechst to label DNA (blue). G Endogenously tagged IMC1 (C-terminal mScarlet-I) to visualize daughter cell formation and ectopically expressed H2B-BFP to visualize nuclear dynamics. Schematic of cell division shown for reference. Selected frames from live-cell imaging of TgSNF2b knockdown parasites imaged 48–72 h after auxin-induced knockdown. Phenotypes highlighted by red triangles: nuclear division without cytokinesis (top) and failure of nuclear retention (bottom). Two SWI/SNF chromatin remodeling complexes are defined by their different helicases, TgSNF2a or TgSNF2b. A Co-IP MS of endogenously TgSWI3-mNG, TgSNF2a-2×ALFA, and TgSNF2b-2×ALFA. Plotted abundances of proteins identified with >2 unique peptides for n = 2 or 3 biological replicates for TgSWI3 or TgSNF2a/b, respectively. Significantly enriched proteins are colored according to their Benjamini–Hochberg corrected two-way ANOVA p-values from abundance ratio analysis. B Heatmap of proteins co-purified with TgSWI3, TgSNF2a, or TgSNF2b in at least two pulldowns, based on analysis in (A). Tiles colored by their adjusted p-value in blue or their conservation in representative genomes, the latter reports the lowest e-value found. C Schematic of domain architectures from InterProScan37. HsBRG1 and HsBRM are shown for reference, alongside AlphaFold338 predictions of the DEXDc and HELICc domains from TgSNF2a (orange) and TgSNF2b (blue); overall RMSD of 5.45 Å and domain-specific RMSDs of 2.42 Å or 0.54 Å for DEXDc or HELICc, respectively. D Immunoblot analysis of stability during reciprocal knockdown of TgSNF2a-mAID or TgSNF2b-mAID for increasing lengths of time, representative of two independent experiments. Blots probed with anti-ALFA to monitor the abundance of the non-degraded helicase, anti-V5 to confirm the knockdown, and anti-SAG1 as a loading control. E Competition assays of TgSWI3, TgSNF2a, and TgSNF2b mAID-tagged strains comparing the relative growth against the parental strain expressing IMC1-tdTomato24 in auxin-containing media. At each time point, the proportion of fluorescent parasites was normalized to the vehicle control. F Immunofluorescence images of TgSNF2a and TgSNF2b knockdown parasites following 72 h of treatment, representative of at least two independent experiments. Fixed parasites were stained for GAP45 (magenta), IMC1 (yellow), mNG (green), and Hoechst to label DNA (blue). G Endogenously tagged IMC1 (C-terminal mScarlet-I) to visualize daughter cell formation and ectopically expressed H2B-BFP to visualize nuclear dynamics. Schematic of cell division shown for reference. Selected frames from live-cell imaging of TgSNF2b knockdown parasites imaged 48–72 h after auxin-induced knockdown. Phenotypes highlighted by red triangles: nuclear division without cytokinesis (top) and failure of nuclear retention (bottom). Several proteins identified in the co-IPs were unexpected constituents of the putative SWI/SNF complexes. TGME49_229790, TGME49_297690, and TGME49_233160 lacked homologs beyond the closest relatives of Toxoplasma, suggesting parasite-specific adaptations of the SWI/SNF complex (Fig. 3B). TGME49_320300 and TgSWI3 co-IP’d with the Myb protein TGME49_203950 (TgTFIIIB). However, interactions between SWI/SNF complexes and TFIIIB have not been previously reported, and TFIIIB is not typically associated with HMG domains21,35. Closer inspection revealed that all peptides identified in the TgSWI3 pulldown mapped exclusively to the HMG domain–containing N-terminus of the gene model, with no coverage of the annotated C-terminal Myb domain. The uneven distribution of RNA-seq reads across the locus, peaking in the middle of the annotated gene model, further suggest the current annotation is inaccurate and the locus produces two distinct transcripts (Supplementary Fig. 4A). Experimental validation confirmed that TGME49_203950 indeed gives rise to two distinct transcripts, encoding separate proteins that we will refer to as TgHMGB3 for the N-terminal portion and TgTFIIIB for the C-terminal portion (Supplementary Fig. 4B–D), and only TgHMGB3 interacts with the TgSWI3 complex. The two helicases (TGME49_278440 and TGME49_320300) stand out as the best-conserved elements across the analyzed eukaryotic lineages (Fig. 3B). Phylogenetic analysis of the 17 most closely related (e-value cutoff: 0.01) Toxoplasma proteins containing both DEXDc and HELICc domains revealed distinct clades with homology to canonical chromatin remodeler families, including CHD, ISWI, and INO80, as well as RAD-like ATPases involved in DNA repair (Supplementary Fig. 5A). The two helicases identified in the TgSWI3 co-IPs were closely related members of the SWI/SNF clade (Supplementary Fig. 5A–C). Others have suggested these two proteins are candidate SWI/SNF helicases and designated them TgSNF2a (TGME49_278440) and TgSNF2b (TGME49_320300), additionally noting that Plasmodium has a single helicase of this type8,17,36. Alignment with SWI/SNF helicases from other eukaryotes identified several hallmark SWI/SNF domains, with HSA and SnAC domains uniquely present in TgSNF2b, and the BROMO domain specific to TgSNF2a. These domain assignments were further supported by InterProScan37 predictions (Fig. 3C, Supplementary Fig. 5B, and Supplementary Table 1). Although humans similarly have two such helicases—HsBRG1 and HsBRM—we were not able to confidently pair either with the Toxoplasma proteins. Furthermore, TgSNF2a and TgSNF2b are predicted to adopt similar overall folds within their ATPase cores, despite notable differences in their predicted structures (AlphaFold338 RMSD: 5.45 Å) and primary sequence (42.18 % identity; Fig. 3C). We considered whether depletion of one helicase might destabilize or co-deplete the other34,39. To test this, we generated dually-tagged parasite lines in which either TgSNF2a or TgSNF2b was endogenously fused to an mAID, while the reciprocal helicase was tagged with a 2×ALFA epitope, enabling depletion of one helicase and monitoring of the other. We assayed TgSNF2a and TgSNF2b abundance following different periods of knockdown. Immunoblot analysis showed that depletion of either TgSNF2a or TgSNF2b did not impact the stability or abundance of the other. These results indicate that the two SWI/SNF subcomplexes remain independently assembled and stable upon loss of the alternate helicase (Fig. 3D; Source Data), consistent with the formation of discrete TgSWI3-associated chromatin remodeling complexes incorporating either TgSNF2a or TgSNF2b. Functional divergence between TgSNF2a and TgSNF2b reveals non-redundant roles in Toxoplasma We assessed the individual contributions of TgSNF2a and TgSNF2b to parasite fitness by performing competition assays in the presence of auxin. These experiments revealed that TgSNF2a-depleted parasites were rapidly outcompeted by the wildtype parental line with similar kinetics to TgSWI3-depleted parasites24. By contrast, TgSNF2b-depleted parasites displayed a comparatively milder phenotype, suggesting a more acute functional requirement for TgSNF2a during the Toxoplasma lytic cycle (Fig. 3E and Supplementary Fig. 6A; Source Data). To further dissect the functional consequences of depleting the two SWI/SNF helicases, we performed immunofluorescence microscopy following 72 h of TgSNF2a or TgSNF2b knockdown. Strikingly, depletion of TgSNF2a resulted in severe cell division defects, characterized by aberrant nuclear morphology and pronounced changes in Hoechst staining intensity, indicative of altered chromatin structure and compaction40. By contrast, TgSNF2b knockdown led to the emergence of parasites appearing to undergo endopolygeny, with more than two daughter cells forming within a single mother cell41. This phenotype was observed in ~30 % of parasite vacuoles and often affected only a subset of parasites within the same vacuole, suggesting dysregulation of the parasite’s cell cycle (Fig. 3F and Supplementary Fig. 6B; Source Data). To observe the cell division phenotype associated with TgSNF2b depletion by live-cell imaging, we generated a conditional TgSNF2b knockdown strain expressing fluorescently tagged markers for the inner membrane complex (IMC1) and chromatin (H2B; Fig. 3G). Time-lapse microscopy showed that TgSNF2b knockdown disrupts the coordination between mitosis and cytokinesis. In a subset of parasites, nuclear division proceeds but is not followed by cytokinesis, leading to the accumulation of multinucleated cells and subsequent formation of multiple daughters within a single mother cell. In other cases, daughter cells fail to retain their nucleus during cytokinesis, resulting in the presence of nuclei within the residual body (Fig. 3G and Supplementary Movie 1). These observations indicate that TgSNF2b is critical for the faithful progression of nuclear segregation and cytokinesis during parasite replication. TgSNF2a and TgSNF2b depletion have distinct transcriptional outcomes The distinct phenotypes resulting from TgSNF2a and TgSNF2b depletion suggest that the associated complexes contribute differentially to gene expression. To investigate their transcriptional impact, we performed transcriptional profiling following knockdown for 6, 12, or 24 h. Loss of either helicase led to up- and down-regulation of transcripts, with downregulated genes constituting a larger proportion of the response. For example, after 12 h of knockdown, TgSNF2a depletion resulted in 201 downregulated and 151 upregulated genes, whereas TgSNF2b depletion yielded 62 downregulated and 30 upregulated transcripts. The genes upregulated following TgSNF2a depletion were largely lowly expressed under baseline conditions (normalized counts <37.9) and thus prone to high variance; accordingly, they were excluded from subsequent analyses across the TgSWI3, TgSNF2a, and TgSNF2b datasets (Fig. 4A and Supplementary Fig. 7A). Overall, TgSNF2a knockdown led to a faster and more substantial transcriptional disruption compared to TgSNF2b, underscoring the immediate dependence of gene expression on TgSNF2a (Fig. 4A and Supplementary Data 4). PCA revealed that TgSNF2a and TgSNF2b knockdowns produce markedly different transcriptional outcomes, with TgSWI3 occupying an intermediate position between the two, reflecting a mixed relationship. Moreover, TgSNF2b knockdown showed transcriptional progression over time, while TgSNF2a and TgSWI3 profiles remained relatively stable (Fig. 4B and Supplementary Fig. 7B). Together, these findings suggest that TgSNF2a and TgSNF2b operate in distinct temporal regimes, while TgSWI3 integrates features of both complexes, reflecting its role as a scaffold in the two SWI/SNF complexes.Fig. 4Time-resolved transcriptomic analysis reveals distinct regulatory roles for TgSNF2a and TgSNF2b.A Plot displaying the number of significantly upregulated and downregulated genes following knockdown of TgSNF2a, TgSNF2b, or TgSWI3, as determined by DESeq278. Differentially expressed genes were defined by an adjusted p-value below 0.005 and an absolute fold change greater than 2. The proportion of genes with low read counts (<37.9) is annotated. B PCA performed on DESeq278 variance-stabilized expression values generated in biological triplicates for each condition and time point, using genes with a variance >0.1 across all samples. Genes with low read counts across all samples were excluded. The first two principal components are shown. C Venn diagrams of the overlap of up- and downregulated genes following a 12-h knockdown of TgSNF2a (orange), TgSNF2b (blue), or TgSWI3 (red). D Expression trajectories of genes classified as downregulated after the 12-h knockdown of TgSNF2a, TgSNF2b, or TgSWI3. Lines represent the mean log2 fold change at each time point; shaded areas indicate s.e.m. Significance at the latest time point for each factor was evaluated using a one-sided Wilcoxon signed-rank test. E Pairwise comparisons of transcriptomic changes following 12-hr knockdown of TgSNF2a or TgSNF2b, compared to TgSWI3. Scatter plots display log₂ fold changes from RNA-seq data. Genes are color-coded by their cell cycle-regulated classification based on ref. 33. F Schematic of Gini score calculation, illustrated for low (gene 1) and a high (gene 2) Gini score genes. Mean log₂ fold change in RNA abundance per bin was calculated following the 12-h knockdown of TgSNF2a, TgSNF2b, or TgSWI3. G Pairwise comparisons of transcriptomic changes following the 12-h knockdown of TgSNF2a or TgSNF2b, compared to TgSWI3. Scatter plots display log₂ fold changes from RNA-seq data. Genes color-coded by their log₂ fold change in EES1-stage parasites relative to tachyzoites, based on ref. 63. In panels E, G, dot size corresponds to the mean RNA-seq counts in vehicle-treated samples of the corresponding strains. Time-resolved transcriptomic analysis reveals distinct regulatory roles for TgSNF2a and TgSNF2b. A Plot displaying the number of significantly upregulated and downregulated genes following knockdown of TgSNF2a, TgSNF2b, or TgSWI3, as determined by DESeq278. Differentially expressed genes were defined by an adjusted p-value below 0.005 and an absolute fold change greater than 2. The proportion of genes with low read counts (<37.9) is annotated. B PCA performed on DESeq278 variance-stabilized expression values generated in biological triplicates for each condition and time point, using genes with a variance >0.1 across all samples. Genes with low read counts across all samples were excluded. The first two principal components are shown. C Venn diagrams of the overlap of up- and downregulated genes following a 12-h knockdown of TgSNF2a (orange), TgSNF2b (blue), or TgSWI3 (red). D Expression trajectories of genes classified as downregulated after the 12-h knockdown of TgSNF2a, TgSNF2b, or TgSWI3. Lines represent the mean log2 fold change at each time point; shaded areas indicate s.e.m. Significance at the latest time point for each factor was evaluated using a one-sided Wilcoxon signed-rank test. E Pairwise comparisons of transcriptomic changes following 12-hr knockdown of TgSNF2a or TgSNF2b, compared to TgSWI3. Scatter plots display log₂ fold changes from RNA-seq data. Genes are color-coded by their cell cycle-regulated classification based on ref. 33. F Schematic of Gini score calculation, illustrated for low (gene 1) and a high (gene 2) Gini score genes. Mean log₂ fold change in RNA abundance per bin was calculated following the 12-h knockdown of TgSNF2a, TgSNF2b, or TgSWI3. G Pairwise comparisons of transcriptomic changes following the 12-h knockdown of TgSNF2a or TgSNF2b, compared to TgSWI3. Scatter plots display log₂ fold changes from RNA-seq data. Genes color-coded by their log₂ fold change in EES1-stage parasites relative to tachyzoites, based on ref. 63. In panels E, G, dot size corresponds to the mean RNA-seq counts in vehicle-treated samples of the corresponding strains. We further investigated the temporal dynamics and specificity of transcriptional responses to disrupting the different SWI/SNF complexes. Venn diagrams of genes that were up- or down-regulated following 12 h of knockdown for TgSNF2a, TgSNF2b, or TgSWI3 indicated that most differentially expressed genes were associated with the depletion of individual complex subunits (Fig. 4C). This pattern may result from overly conservative thresholds in differential-expression analysis (adjusted p-value < 0.005 and absolute fold change >2), which might not capture weaker trends in gene expression. As an alternative analysis, we examined the behavior of downregulated genes across the various knockdowns after 12 h (Fig. 4D). In TgSNF2a-depleted parasites, genes associated with TgSNF2b-dependent downregulation remained largely unaffected, while the TgSWI3-associated subset showed moderate downregulation. By contrast, depletion of TgSNF2b resulted in the downregulation of genes associated with both TgSNF2a and TgSWI3, although the magnitude of repression was less pronounced than for genes associated with TgSNF2b based on magnitude and significance cutoffs. As expected, TgSWI3 depletion caused downregulation across all three subsets. This pattern suggests an asymmetric codependence, with TgSNF2a-dependent genes requiring TgSNF2b while TgSNF2b-dependent genes are less reliant on TgSNF2a. Together, these findings support a model in which TgSNF2b acts as a general transcriptional activator, while TgSNF2a is selectively required for the activation of a distinct subset of genes that depend more acutely on its presence. Prompted by the distinct cell division defects observed upon TgSNF2a and TgSNF2b depletion, we investigated how cell cycle-regulated genes respond to the loss of these remodelers. We compared transcriptional profiles at 12-h post-knockdown relative to TgSWI3 depletion. Consistent with our earlier observations, TgSNF2b and TgSWI3 knockdowns predominantly affected the same set of genes, though with distinct kinetics, with TgSWI3 depletion triggering faster transcriptional downregulation (Fig. 4E). By contrast, TgSNF2a-sensitive genes were enriched for genes expressed during short-lived cell cycle stages such as mitosis and cytokinesis33. These genes strongly correlated with high Gini scores—a measure of distribution inequality that quantifies variation across samples ranging from genes expressed equally across all stages (Gini score = 0) to genes expressed uniquely at a single stage (Gini score = 1)42. Gini scores were calculated from published transcriptome data obtained from synchronized parasites43 (Fig. 4F and Supplementary Fig. 7C). This pattern suggests that TgSNF2a is more critically required for the transcription of genes with narrow temporal expression windows, such as those activated during mitosis and cytokinesis. Despite their distinct transcriptional profiles, TgSNF2a and TgSNF2b knockdowns share several upregulated genes, many of which are typically repressed in tachyzoites but activated in bradyzoites and early sexual stages (EES1, Fig. 4G and Supplementary Fig. 7C). Similar patterns of de-repression have been observed upon disruption of other ATP-dependent chromatin remodelers, such as TgDMAP1, TgSNF2h (Supplementary Fig. 7D, E)8, and TgMORC7. These findings suggest that while TgSNF2a is required for the transcriptional activation of genes with narrow, phase-restricted expression windows, disruption of chromatin remodeling can lead to de-repression of developmentally silenced transcripts in tachyzoites. Chromatin profiling using nanobody-directed CUT&RUN We sought to define the genome-wide binding profiles of the SWI/SNF complexes. Although chromatin profiling approaches such as ChIP-seq and CUT&RUN have been successfully applied in Toxoplasma, these methods typically require large numbers of parasites and are limited in their ability to capture transient or low-abundance interactions3,6,8,44,45. To overcome these limitations, we developed ALFA-CUT&RUN, a rapid and sensitive chromatin profiling approach optimized for Toxoplasma, extending the original CUT&RUN46,47. Our method leverages the strong and specific interaction between the ALFA epitope tag and its nanobody, NbALFA48. By directly fusing micrococcal nuclease (MNase) to NbALFA, the system eliminates the need for antibodies and enables targeted cleavage of chromatin in proximity to ALFA-tagged proteins. To specifically map the binding sites of the various complexes, parasites expressing C-terminally 2×ALFA-tagged TgSWI3, TgSNF2a, or TgSNF2b were permeabilized, incubated with the NbALFA-MNase fusion protein, and subjected to calcium-dependent nuclease activation. Released DNA fragments were then purified and sequenced to generate high-resolution binding profiles of TgSWI3, TgSNF2a, and TgSNF2b across the genome for two biological replicates (Fig. 5A and Supplementary Fig. 8). To ensure specificity and assess background signal, we included two controls: the untagged parental strain, and a 2×ALFA-tagged TgTFIIIB line, which served as a positive control based on its well-characterized genomic distribution at short and non-coding RNAs in other systems49. This allowed us to distinguish specific enrichment patterns associated with TgSWI3, TgSNF2a, and TgSNF2b from background cleavage or nonspecific interactions. Although we found that all SWI/SNF subunits bound several TgTFIIIB target sites—highlighting their broad role in transcriptional regulation—these sites were excluded from downstream analyses so as to take full advantage of this control (Supplementary Fig. 8D–F). Following stringent differential-binding analysis, the signal intensities from biological replicates for TgSWI3, TgSNF2a and TgSNF2b were strongly correlated (Fig. 5B, C and Supplementary Fig. 8B–F). These results underscore the robustness and specificity of the ALFA-CUT&RUN approach for profiling chromatin-associated proteins in Toxoplasma.Fig. 5Genome-wide profiling of TgSNF2a and TgSNF2b occupancy reveals distinct promoter-binding patterns associated with developmental gene regulation.A Overview of the ALFA-CUT&RUN protocol. B Number of peaks identified per ALFA-CUT&RUN replicate for TgSWI3, TgSNF2a, or TgSNF2b. Bars represent the total number of peaks called in each replicate, with replicated peaks indicated in a darker shade according to IDR scores. C Correlation heatmap of the ALFA-CUT&RUN signal across all samples for 4692 consensus loci. Pairwise Pearson correlation coefficients calculated using normalized and background-subtracted read counts. D Relative enrichment of TgSWI3, TgSNF2a, or TgSNF2b within ±500 bp of annotated TSSs. Nucleosome positions from refs. 7,87 are highlighted. E Distribution of TgSWI3 (red), TgSNF2a (orange), and TgSNF2b (blue) peaks ordered according to average peak intensities, relative to the nearest TSS. Heatmaps display associated transcriptomic data: mean RNA expression in tachyzoites (greyscale); log₂ fold changes across cell cycle phases relative to tachyzoite mean expression, based on ref. 43 (cell cycle stages indicated above with S, M, and C in shades of purple, and G1a and G1b in shades of green); and log2 fold changes in gene expression for tissue cysts and enteroepithelial stages (EES1–5) relative to tachyzoites, based on ref. 63. F–H Comparison of TgSNF2a and TgSNF2b log₂ abundance based on normalized read counts after background signal subtraction. Each point represents a single genomic consensus locus of the 4433 high confidence peak-gene pairs. Points are colored based on mean gene expression level in tachyzoites (F), by the log₂ fold change in expression of EES1 stages relative to tachyzoites based on ref. 63 (G), or by TgMORC fold enrichment based on ref. 7 (H). Genome-wide profiling of TgSNF2a and TgSNF2b occupancy reveals distinct promoter-binding patterns associated with developmental gene regulation. A Overview of the ALFA-CUT&RUN protocol. B Number of peaks identified per ALFA-CUT&RUN replicate for TgSWI3, TgSNF2a, or TgSNF2b. Bars represent the total number of peaks called in each replicate, with replicated peaks indicated in a darker shade according to IDR scores. C Correlation heatmap of the ALFA-CUT&RUN signal across all samples for 4692 consensus loci. Pairwise Pearson correlation coefficients calculated using normalized and background-subtracted read counts. D Relative enrichment of TgSWI3, TgSNF2a, or TgSNF2b within ±500 bp of annotated TSSs. Nucleosome positions from refs. 7,87 are highlighted. E Distribution of TgSWI3 (red), TgSNF2a (orange), and TgSNF2b (blue) peaks ordered according to average peak intensities, relative to the nearest TSS. Heatmaps display associated transcriptomic data: mean RNA expression in tachyzoites (greyscale); log₂ fold changes across cell cycle phases relative to tachyzoite mean expression, based on ref. 43 (cell cycle stages indicated above with S, M, and C in shades of purple, and G1a and G1b in shades of green); and log2 fold changes in gene expression for tissue cysts and enteroepithelial stages (EES1–5) relative to tachyzoites, based on ref. 63. F–H Comparison of TgSNF2a and TgSNF2b log₂ abundance based on normalized read counts after background signal subtraction. Each point represents a single genomic consensus locus of the 4433 high confidence peak-gene pairs. Points are colored based on mean gene expression level in tachyzoites (F), by the log₂ fold change in expression of EES1 stages relative to tachyzoites based on ref. 63 (G), or by TgMORC fold enrichment based on ref. 7 (H). TgSNF2a and TgSNF2b display binding patterns with distinct cell cycle and chromatin signatures We applied the Irreproducible Discovery Rate (IDR) framework to identify loci consistently enriched across biological replicates50. Using an IDR p-value < 0.05, we identified 3,516 reproducible peaks for TgSWI3, 3,229 for TgSNF2a, and 3,823 for TgSNF2b (Fig. 5B, Supplementary Fig. 8A, D–F, and Supplementary Data 5). Analysis of the genomic distribution of replicated peaks revealed that binding sites for TgSWI3, TgSNF2a, and TgSNF2b were enriched at transcription start sites (TSSs; Fig. 5D, E). Nucleosome-level mapping further uncovered distinct binding patterns: TgSNF2a was predominantly centered around the +1 and +2 nucleosomes downstream of the TSS, whereas TgSNF2b was enriched at the −1 and +1 nucleosomes, flanking the nucleosome-depleted region (NDR; Fig. 5D). These findings suggest that the two SWI/SNF complexes engage with promoter architecture in distinct ways, potentially reflecting their specialized functions in gene regulation. To enable the direct comparison of TgSNF2a and TgSNF2b binding across the genome, we defined a set of 4692 consensus regions, representing all loci identified by an IDR p value <0.05 in any of the samples. Of these, 4433 could be confidently assigned to a single promoter, providing a unified set of loci for quantitative analysis of differential binding (Fig. 5E and Supplementary Fig. 8A). Consistent with a direct regulatory role, genes whose promoters were bound by TgSWI3, TgSNF2a, or TgSNF2b showed a significant enrichment among factor-dependent differentially expressed genes (adjusted p-value < 0.05), as assessed by Fisher’s exact test (TgSWI3: p = 2.92e−4; TgSNF2a: p = 1.34e−8; TgSNF2b: p = 3.86e−2). Correlating TgSNF2a and TgSNF2b binding profiles with transcriptomic data across the cell cycle and life cycle stages revealed distinct regulatory patterns. Genes with the highest TgSNF2a and TgSNF2b occupancy were predominantly associated with G1a and G1b phases, whereas loci bound at lower intensity were more frequently linked to S, M, or C phase genes, which are often transiently expressed and require tight temporal control. Moreover, peak intensity was positively correlated with mRNA abundance, suggesting that SWI/SNF occupancy is linked to transcriptional output (Fig. 5E, F and Supplementary Data 6); a conclusion further supported by the trend of downregulation following knockdown of the factors. TgSNF2b bound several loci with reduced or absent TgSNF2a binding (Fig. 5E and Supplementary Fig. 8G, H). These TgSNF2b-enriched peaks were linked to genes that are transcriptionally silent during the lytic cycle but become strongly induced during chronic- and early sexual-stage differentiation, as well as in extracellular parasites (Fig. 5E, G and Supplementary Fig. 9A, B). One such differentiation-associated gene is CST10, which has been identified as an early chronic stage marker that is expressed before other chronic stage markers such as BAG1 or CST13. This suggests that while TgSNF2b is not sufficient on its own to drive transcriptional activation, it maintains these loci in a poised state for rapid activation during developmental transitions. Given the association of TgSNF2b-enriched loci with genes activated during developmental transitions, we compared TgSNF2b and TgSNF2a binding patterns to TgMORC, a transcriptional repressor involved in silencing stage-specific genes (Fig. 5H and Supplementary Fig. 9C, D)7. TgSNF2b was strongly correlated with TgMORC occupancy, whereas TgSNF2a exhibited no such correlation. Genomic profiling further revealed that TgMORC, like TgSNF2b, is predominantly localized at the –1 nucleosome upstream of the transcription start site, consistent with a role in modulating promoter accessibility (Supplementary Fig. 9D). Notably, a subset of genomic regions displayed high levels of TgSNF2a, TgSNF2b and TgMORC co-occupancy (Fig. 5H). These loci were associated with genes that are highly expressed in tachyzoites at the population level, indicating that the functional interplay between TgSNF2a, TgSNF2b, and TgMORC extends beyond developmental regulation and also contributes to transcriptional control in tachyzoites. Late cell cycle stages exhibit an increased dependence on TgSNF2a As noted above, TgSNF2a and TgSNF2b occupancy was higher for genes expressed during G1 than for genes expressed during the S, M, or C phases of the cell cycle. If TgSNF2a and TgSNF2b occupancy changes dynamically across the cell cycle, the observed pattern would be dominated by the profile of the abundant G1 stages, which represent 50 % of parasites in an asynchronous population33. Therefore, we examined how TgSNF2a- and TgSNF2b-binding patterns correlate with the timing and precision of cell cycle-regulated transcription. TgSNF2b was significantly more abundant than TgSNF2a at genes associated with shorter cell cycle phases (Fig. 6A), such as S, M, and C phases, which represent 28, 15, and 7% of asynchronous populations, respectively33. A similar pattern was observed for TgMORC (Supplementary Fig. 10A). By contrast, TgSNF2a abundance closely matched that of TgSNF2b at genes expressed during G1a and G1b. However, starting in S phase and continuing through C phase, TgSNF2b levels were significantly higher than TgSNF2a, suggesting a broader or more prolonged association of TgSNF2b with these loci across the parasite population. This differential enrichment may indicate that TgSNF2b plays a more sustained role in maintaining chromatin accessibility at these loci, whereas TgSNF2a may be required only transiently (Fig. 6A, B).Fig. 6TgSNF2a and TgSNF2b remodel chromatin with distinct temporal roles during the Toxoplasma cell cycle.A Factor abundance at phase-specific promoters, classified based on ref. 33. Differences assessed by two-way Wilcoxon rank-sum test. Boxplots show the median and 25th–75th percentiles (box) and whiskers extending to 1.5×IQR for n = 43–263 genes per category. B GSEA of cell cycle gene sets. NES indicates whether TgSNF2a or TgSNF2b is preferentially associated; p values calculation described in the Methods. C Mean log₂ abundance of TgSNF2a and TgSNF2b across Gini score bins (size = 0.005), from synchronized parasite transcriptomes43; gene count (top panel); p value (bottom panel) represents the significance of the slope difference between TgSNF2a and TgSNF2b by weighted linear regression of abundance as function of Gini score. Right panel highlights cell cycle-regulated genes (ref. 33); gene count is shown in light gray, and mean log₂ abundance of TgSNF2a or TgSNF2b in colored plots. D, E Nuclear mNG fluorescence from live-cell video microscopy of endogenously tagged TgSNF2a (D) and TgSNF2b (E). Cell cycle stages are annotated above the plot. Gray bars mark periods of cytokinesis. Individual biological replicates (n = 4) are shown as gray lines, with the mean in orange or blue. Expected photobleaching (green line) shown with 95% confidence interval (shaded area). F Replication in the presence or absence of auxin after 24 h of growth. Plotted mean ± s.d for n = 3 biological replicates. G Analysis of DNA content of TgSNF2a knockdown strains after increasing knockdown periods. Representative result from three independent experiments. Line shows the gate segregating G1 (1 N) and replicating (S, M, C: >1–2 N) parasites. Percentages represent the mean proportion of S/M/C-phase parasites across n = 3 biological replicates. H DNA content staining of parental or TgSNF2a and TgSNF2b knockdown strains after different treatment periods (as in G). Dotted line indicates the wildtype value (50%)33. Significance assessed using Welch’s one-tailed t-tests. I Live-cell imaging of TgSNF2a knockdown parasites expressing IMC1-mScarlet (magenta) and H2B-BFP (white). Parasites were imaged over a 6–30 h after auxin treatment. Selected frames shown from 6 to 30 h post-knockdown. TgSNF2a and TgSNF2b remodel chromatin with distinct temporal roles during the Toxoplasma cell cycle. A Factor abundance at phase-specific promoters, classified based on ref. 33. Differences assessed by two-way Wilcoxon rank-sum test. Boxplots show the median and 25th–75th percentiles (box) and whiskers extending to 1.5×IQR for n = 43–263 genes per category. B GSEA of cell cycle gene sets. NES indicates whether TgSNF2a or TgSNF2b is preferentially associated; p values calculation described in the Methods. C Mean log₂ abundance of TgSNF2a and TgSNF2b across Gini score bins (size = 0.005), from synchronized parasite transcriptomes43; gene count (top panel); p value (bottom panel) represents the significance of the slope difference between TgSNF2a and TgSNF2b by weighted linear regression of abundance as function of Gini score. Right panel highlights cell cycle-regulated genes (ref. 33); gene count is shown in light gray, and mean log₂ abundance of TgSNF2a or TgSNF2b in colored plots. D, E Nuclear mNG fluorescence from live-cell video microscopy of endogenously tagged TgSNF2a (D) and TgSNF2b (E). Cell cycle stages are annotated above the plot. Gray bars mark periods of cytokinesis. Individual biological replicates (n = 4) are shown as gray lines, with the mean in orange or blue. Expected photobleaching (green line) shown with 95% confidence interval (shaded area). F Replication in the presence or absence of auxin after 24 h of growth. Plotted mean ± s.d for n = 3 biological replicates. G Analysis of DNA content of TgSNF2a knockdown strains after increasing knockdown periods. Representative result from three independent experiments. Line shows the gate segregating G1 (1 N) and replicating (S, M, C: >1–2 N) parasites. Percentages represent the mean proportion of S/M/C-phase parasites across n = 3 biological replicates. H DNA content staining of parental or TgSNF2a and TgSNF2b knockdown strains after different treatment periods (as in G). Dotted line indicates the wildtype value (50%)33. Significance assessed using Welch’s one-tailed t-tests. I Live-cell imaging of TgSNF2a knockdown parasites expressing IMC1-mScarlet (magenta) and H2B-BFP (white). Parasites were imaged over a 6–30 h after auxin treatment. Selected frames shown from 6 to 30 h post-knockdown. The classification of genes to a specific cell cycle phase relies on peak expression; however, it does not consider how much of the total expression is stage-specific. To capture the stage specificity of expression, we again used the Gini score, where higher values indicate transcripts that are predominantly expressed in one cell cycle phase and likely repressed in others. TgSNF2b occupancy increases with higher Gini scores, indicating a possible link with genes that remain poised for most of the cell cycle (Fig. 6C and Supplementary Fig. 10B, C)43. These results reinforce the notion that TgSNF2a and TgSNF2b are associated with different patterns of gene regulation, and TgSNF2b is particularly associated with poised or repressed loci not only in preparation for differentiation but also throughout the Toxoplasma cell cycle. The association of TgSNF2a and TgSNF2b with promoters of cell cycle-regulated genes prompted us to investigate whether their abundance changes across the cell cycle. We measured the intensity of mNG endogenously fused to either TgSNF2a or TgSNF2b, using live video microscopy. TgSNF2a levels changed modestly throughout the cell cycle. However, TgSNF2b levels declined following cytokinesis and remained low during G1a and G1b, followed by a sharp increase during S phase that coincided with the onset of DNA replication (Fig. 6D, E and Supplementary Data 7 and 8). This S-phase–specific surge in TgSNF2b abundance suggests a role in preserving promoter accessibility during replication-coupled chromatin assembly, in contrast to the stable levels of TgSNF2a, which are consistent with the proposed transient role during all transcriptional activation events across the cell cycle. Knockdown of TgSNF2a, but not TgSNF2b, resulted in a significant reduction in parasite numbers per vacuole, consistent with an immediate slowdown of the parasite's cell cycle (Fig. 6F; Source Data). To determine whether this defect was linked to a specific cell cycle phase, we performed propidium iodide staining to assess DNA content following 12, 24, and 48 h of knockdown (Fig. 6G, H; Source Data). TgSNF2a-depleted parasites exhibited an increased proportion of cells in S, M, and C phases, a shift not observed after TgSNF2b knockdown. These results indicate that TgSNF2a is required for timely progression through DNA replication and mitosis. Consistent with this, live-cell imaging revealed prolonged cell cycle phases in TgSNF2a-depleted parasites, during which mitosis frequently preceded the formation of daughter cells by ~1 h (Fig. 6I and Supplementary Movie 2)—a phenotype that was not observed in the vehicle-treated control, where daughter cell budding occurred concurrently with mitosis (Supplementary Fig. 10D and Supplementary Movie 3). Together, these findings implicate TgSNF2a-dependent chromatin remodeling in the regulation of late cell cycle transitions and the maintenance of proper division timing, highlighting its importance for rapid transcriptional activation during short, transcriptionally dynamic phases of the cell cycle. Dynamic binding of TgSNF2a and TgSNF2b across the cell cycle is shaped by chromatin context To directly address the possibility that SWI/SNF chromatin interactions vary across the cell cycle, we performed cell cycle–resolved ALFA-CUT&RUN of TgSNF2a and TgSNF2b. Therefore, parasites were synchronized using a thymidine-synchrony model51, and harvested at defined stages corresponding to G1, early S, mid S and M/C phases (Fig. 7A and Supplementary Data 9). We quantified the occupancy of TgSNF2a and TgSNF2b at all 4,433 high-confidence loci and compared their enrichment at previously classified cell cycle stage-specific genes33. This analysis revealed distinct temporal binding patterns that closely mirrored the transcriptional dynamics throughout the cell cycle. For example, promoters of G1-specific genes displayed a higher TgSNF2a/TgSNF2b ratio in G1-stage parasites, consistent with the association of TgSNF2a at transcriptionally active loci. By contrast, this ratio flipped in S and M/C phases, indicating a relative increase in TgSNF2b occupancy or loss of TgSNF2a at those G1-specific promoters (Fig. 7B). Although early S, mid S, and M/C phases displayed broadly similar profiles, a temporal trend remained apparent with some G1 genes exhibiting TgSNF2a enrichment as early as the M/C stage (Fig. 7B). Together, these observations support a model in which TgSNF2a is dynamically and transiently recruited to promoters in a phase-specific manner, whereas TgSNF2b persists or accumulates more strongly at poised promoters.Fig. 7Cell cycle–resolved CUT&RUN uncovers distinct, phase-specific and chromatin-associated occupancy patterns for TgSNF2a and TgSNF2b.A Schematic representation of cell cycle synchronization using thymidine (Th.) containing media (dark gray) followed by washouts (light gray). Bottom panels display representative flow cytometry profiles of the DNA content in synchronized parasites used for cell cycle–resolved CUT&RUN, assessed by propidium iodide staining. B Heatmap showing the TgSNF2b/TgSNF2a ratio at four cell cycle stages. High-confidence consensus peaks of genes found to have cell cycle-specific expression patterns by ref. 33 are shown, with categories indicated in the bottom row. Z-score normalized log₂ fold changes were hierarchically clustered using complete linkage on Euclidean distances. C Correlation heatmap of indicated chromatin-associated features across all genes associated with a consensus locus between TgSWI3, TgSNF2a, and TgSNF2b samples. Pairwise Pearson correlation coefficients were calculated using log2-transformed mean-centered values. Light gray bars indicate that the signal was measured over the promoter region within the SWI/SNF consensus locus, whereas dark gray bars denote that the signal was assessed across the gene body. Chromatin-associated features were derived from refs. 7,44,52. D Multiple linear regression models were fit to assess the contribution of chromatin-associated features to transcriptional changes following TgSNF2a, TgSNF2b, or TgSWI3 depletion. Shown are regression coefficients (estimated log2 fold change) for each chromatin feature, with 95% confidence intervals (gray bars). The point size reflects the statistical significance of each predictor, indicating the relative strength of association. E Schematic model illustrating a proposed mechanism of transcriptional regulation mediated by TgSNF2a or TgSNF2b associated SWI/SNF complexes. Genes poised in a repressed chromatin state are characterized by H4K31ac enrichment, TgSNF2b, and TgMORC occupancy. Transition to an active state involves loss of TgMORC, acquisition of H3K14ac, and recruitment of TgSNF2a. This chromatin remodeling cascade facilitates RNA polymerase II or RNA polymerase III-based transcriptional activation. Cell cycle–resolved CUT&RUN uncovers distinct, phase-specific and chromatin-associated occupancy patterns for TgSNF2a and TgSNF2b. A Schematic representation of cell cycle synchronization using thymidine (Th.) containing media (dark gray) followed by washouts (light gray). Bottom panels display representative flow cytometry profiles of the DNA content in synchronized parasites used for cell cycle–resolved CUT&RUN, assessed by propidium iodide staining. B Heatmap showing the TgSNF2b/TgSNF2a ratio at four cell cycle stages. High-confidence consensus peaks of genes found to have cell cycle-specific expression patterns by ref. 33 are shown, with categories indicated in the bottom row. Z-score normalized log₂ fold changes were hierarchically clustered using complete linkage on Euclidean distances. C Correlation heatmap of indicated chromatin-associated features across all genes associated with a consensus locus between TgSWI3, TgSNF2a, and TgSNF2b samples. Pairwise Pearson correlation coefficients were calculated using log2-transformed mean-centered values. Light gray bars indicate that the signal was measured over the promoter region within the SWI/SNF consensus locus, whereas dark gray bars denote that the signal was assessed across the gene body. Chromatin-associated features were derived from refs. 7,44,52. D Multiple linear regression models were fit to assess the contribution of chromatin-associated features to transcriptional changes following TgSNF2a, TgSNF2b, or TgSWI3 depletion. Shown are regression coefficients (estimated log2 fold change) for each chromatin feature, with 95% confidence intervals (gray bars). The point size reflects the statistical significance of each predictor, indicating the relative strength of association. E Schematic model illustrating a proposed mechanism of transcriptional regulation mediated by TgSNF2a or TgSNF2b associated SWI/SNF complexes. Genes poised in a repressed chromatin state are characterized by H4K31ac enrichment, TgSNF2b, and TgMORC occupancy. Transition to an active state involves loss of TgMORC, acquisition of H3K14ac, and recruitment of TgSNF2a. This chromatin remodeling cascade facilitates RNA polymerase II or RNA polymerase III-based transcriptional activation. To better understand the chromatin context of bound promoters and the associated gene bodies, we integrated chromatin profiles of published histone modifications, histone variants, and chromatin-associated factors, computing Pearson correlations between the occupancy profiles of TgSNF2a, TgSNF2b, and TgSWI3 and the various chromatin features7,44,52. Consistent with our earlier observations of co-occupancy between TgSNF2b and TgMORC (Fig. 5H), TgSNF2b binding was strongly correlated with binding of TgHDAC3, a known TgMORC-interacting histone deacetylase (Fig. 7C and Supplementary Fig. 11A)7. By contrast, TgSNF2a was strongly correlated with H4K31ac and H3K14ac, which are typically associated with increased chromatin accessibility52. Although TgSNF2b was also positively correlated with H4K31ac, its presence was negatively correlated with H3K14ac, suggesting divergent interactions with acetylated chromatin. Notably, the active promoter mark H3K4me3 was positively associated with TgSNF2a but negatively correlated with TgSNF2b. The opposite trend was observed for the repressive H3K9me3 gene body mark52, which was positively correlated with TgSNF2b but negatively correlated with TgSNF2a. A weak positive correlation was observed between the SWI/SNF factors and gene body H3.3, a histone variant that marks active and poised promoters. No correlation was found with H2AX, which is associated with gene silencing during the DNA damage response and with silenced promoters activated during chronic stage differentiation (Fig. 7C, Supplementary Fig. 11A; and Supplementary Data 10)53–56. These patterns highlight the distinct chromatin environments associated with TgSNF2a and TgSNF2b, where TgSNF2a preferentially associates with open chromatin marks such as histone acetylation, while TgSNF2b is predominantly colocalized with repressive chromatin features such as reduced histone acetylation and enrichment of silencing complexes. We turned to multiple linear regression modeling to determine if these chromatin features can explain transcriptional outcomes following SWI/SNF disruption. The modeling estimates the change in gene expression following knockdown of TgSNF2a, TgSNF2b, or TgSWI3 as a function of increasing a given chromatin feature by one standard deviation, while holding all other variables constant. For example, increased abundance of H3K4me3 is associated with upregulation upon TgSNF2b knockdown, whereas its association with TgSNF2a-dependent expression was comparatively weak (Fig. 7D). H3K14ac and H4K31ac displayed distinct patterns across SWI/SNF perturbations. High H3K14ac levels were predictive of upregulation following TgSNF2b and TgSWI3 depletion, but had little effect on TgSNF2a-dependent transcription. By contrast, H4K31ac consistently predicted repression across the knockdowns of all three factors, with the strongest effects observed for TgSWI3, which is part of both TgSNF2a and TgSNF2b complexes (Fig. 7D). Repression was additionally associated with elevated H2AX levels within the gene body. Therefore, despite the lack of correlation between TgSNF2b and H2AX binding, genes highly bound by H2AX are more likely to be downregulated following TgSNF2b knockdown. A striking divergence was observed between TgMORC and TgSNF2b abundance. High TgMORC levels predicted downregulation upon TgSNF2a loss, but instead predicted upregulation for TgSNF2b and TgSWI3 knockdowns. In addition, high TgSNF2b abundance in promoter regions was associated with upregulation following TgSNF2b depletion but had a repressive effect after TgSNF2a knockdown (Fig. 7D). These results suggest that the complex pattern observed following knockdown of the two SWI/SNF complexes is shaped by the chromatin context of each locus. Discussion Transcriptional regulation coordinates proliferation and developmental transitions in apicomplexan parasites1–8. Numerous sequence-specific transcription factors have been identified as critical regulators of cell cycle progression and differentiation13,18. However, while these factors have well-defined roles in gene expression control, the mechanisms by which chromatin is restructured for transcriptional activation or repression remain largely unexplored. In this study, we systematically examined the family of Myb domain–containing proteins, since apicomplexans have preferentially retained SANT-type Myb proteins that are commonly found in chromatin remodeling complexes. Using high-throughput tagging24 we comprehensively analyzed eight essential factors exhibiting diverse phenotypic and transcriptional effects. We reveal distinct roles during the Toxoplasma cell cycle for factors representing three types of ATP-dependent chromatin remodeling complexes: ISWI, INO80, and SWI/SNF. We focused on the SWI/SNF complex, revealing two mutually exclusive SNF2-type ATPase subunits: TgSNF2a and TgSNF2b. Functional characterization demonstrated that TgSNF2a and TgSNF2b fulfill distinct yet cooperative regulatory roles. TgSNF2a knockdown preferentially impaired the expression of genes active during short cell cycle phases, such as mitosis and cytokinesis, resulting in slowed or stalled progression through these stages. By contrast, TgSNF2b knockdown induced progressive transcriptional repression and disrupted mitotic coordination, leading to multinucleated parasites and abnormal daughter cell formation. Notably, we also found that TgSNF2b co-localizes with the repressor TgMORC at poised promoters and contributes to transcriptional priming of genes activated during chronic and sexual development. Together, these findings suggest a division of labor between the SWI/SNF complexes, with TgSNF2b broadly supporting transcriptional competency, while TgSNF2a mediates transitions in gene expression. By profiling nuclear-localized Myb proteins required for parasite fitness, we identified five candidates whose knockdown led to widespread transcriptional dysregulation and pronounced cell cycle–related phenotypes. Four of these were predicted to function as chromatin regulators based on domain composition and homology, and one corresponded to the conserved spliceosome-associated factor CDC5L. Knockdown of TgCDC5L resulted in cell cycle arrest in G1. While this manuscript was under preparation, a complementary study found that intron retention and the associated formation of protein aggregates likely underlie the cell cycle arrest observed after TgCDC5L knockdown30. We focused on chromatin regulators that exhibit varying degrees of conservation among apicomplexans, which included TgSNF2h, a conserved ATPase characteristic of ISWI-family remodelers, and TgDMAP1, an INO80 cofactor. Knockdown of TgSNF2h and TgDMAP1 induced broadly comparable transcriptional responses, characterized by preferential downregulation of G1-expressed genes and modest upregulation of transcripts expressed during S, M, and C phases, though with contrasting dependencies on TgMORC. TgDMAP1 depletion led to upregulation of transcripts with high TgMORC occupancy, suggesting that TgDMAP1 may cooperate with TgMORC for gene silencing. In canonical systems, DMAP1 is involved in chromatin-regulatory complexes with opposing functions. As part of a silencing complex, DMAP1 recruits DNMT1 and HDAC2 to newly replicated chromatin, facilitating transcriptional repression via DNA methylation and histone deacetylation57. Although Toxoplasma was initially thought to lack DNA methylation, later studies confirmed its presence and identified two functional TgDNMTs58,59. Conversely, within the INO80 SRCAP complex, DMAP1 contributes to transcriptional activation by promoting histone exchange of canonical H2A–H2B dimers with H2A.Z–H2B dimers60. These findings imply that the repressive functions of TgDMAP1 represent the dominant function in tachyzoites. Genes upregulated upon TgSNF2h depletion were generally marked by lower TgMORC levels. In mammals, SNF2h and its paralog SNF2L maintain transcription factor binding by organizing phased nucleosome arrays immediately adjacent to their binding sites. In Toxoplasma, TgSNF2h appears to play an analogous role. It has recently been shown that loss of TgSNF2h triggers widespread chromatin condensation, blocking both TgMORC and transcription factor access at many loci. Yet, its depletion also induces local accessibility shifts that expose new regulatory elements and drive TgMORC-independent gene activation8. SNF2h and SNF2L are thought to be primarily recruited via interactions with transcription factors and, to some extent, by specific histone marks61. TgSNF2h has similarly been shown to interact with TgAP2VIII-2 and TgRFTS, supporting a conserved recruitment mechanism8. Interestingly, in mammalian systems, SNF2h has been found to directly interact with DNMT1 to enhance the deposition of repressive chromatin marks62. While this specific interaction may not be conserved in Toxoplasma, the similar transcriptional and phenotypic consequences of TgSNF2h and TgDMAP1 knockdown suggest that these factors may exert complementary functions. In contrast to the prominent transcriptional dysregulation observed following TgDMAP1 or TgSNF2h depletion, TgSWI3 knockdown resulted in comparatively modest changes in gene expression but pronounced morphological defects that included multinucleation and altered chromatin condensation. These distinct phenotypes suggested a unique role for TgSWI3 in coordinating nuclear division and chromatin organization. In animals, plants, and yeast, SWI3 serves as a core structural component of SWI/SNF remodeling complexes, which generally consist of three functional modules: (1) the base module formed by structural components like SWI3 and subunits involved in binding the nucleosome surface, histone tails and DNA through sequence-agnostic contacts; (2) the motor module composed of a SNF2-type ATP-dependent helicase (BRG1; SMARCA4 or BRM; SMARCA2); and (3) the actin-related protein (ARP) module which connects the base module with the motor module21,34. Through IP and reciprocal co-IP’s, we identified 17 components of the Toxoplasma SWI/SNF complex. These include two mutually exclusive SNF2-type helicases—TgSNF2a and TgSNF2b—as well as five shared core components present in both complexes. All of the shared components have been found to be fitness-conferring in a genome-wide CRISPR screen, with TgSS18 showing the mildest phenotype score at –2.1425. Notably, while both helicases were well conserved across a broad range of representative eukaryotic lineages, many of the remaining complex members showed limited sequence conservation. Among these were canonical core subunits such as SWI3 (SMARCC1/2), SMARCD1, and SMARCB1, which are typically shared between different SWI/SNF subcomplexes or SS18, which is found in specific SWI/SNF subtypes21,34. Additional identified proteins include components that may contribute to the ARP module, such as TgALP2a, or control SWI/SNF stability through ubiquitination, such as RING finger–containing proteins (TGME49_273150 and TGME49_226780)21. Substrate recognition may be mediated by DNA-binding proteins such as TgARID, TgHMGB3 or TgBRD4, a bromodomain-containing protein likely involved in binding acetylated histones34. Together, these findings highlight both conserved and parasite-specific components of the Toxoplasma SWI/SNF complex and underscore substantial adaptations that may reflect unique regulatory demands imposed by the parasite’s divergent chromatin landscape and complex life cycle7,15,16. Additional work will be needed to comprehensively define functional SWI/SNF subcomplexes and to elucidate the specific contributions of novel, apicomplexan-adapted components. Combining transcriptomics and chromatin profiling, we revealed a division of labor between TgSNF2a- and TgSNF2b-containing complexes, which function collaboratively to regulate gene expression in Toxoplasma. Genes with a short phase of expression showed greater TgSNF2b occupancy compared to TgSNF2a, which is explained by the probability of finding each factor on its target locus in an asynchronous parasite population, since based on our model, TgSNF2b would mark such promoters throughout the cell cycle, while TgSNF2a would only associate during active transcription. Notably, we also identified a subset of loci that was bound by TgSNF2b and TgMORC but not by TgSNF2a; accordingly, these genes were not expressed during the lytic cycle7. They were enriched for developmentally regulated genes expressed during parasite differentiation7,63. While this pattern has not been appreciated in other systems, these findings suggest distinct temporal roles for TgSNF2a and TgSNF2b throughout the Toxoplasma cell cycle, with transcription being more acutely dependent on the presence of TgSNF2a while TgSNF2b has a more general function in maintaining chromatin in a poised state for rapid activation throughout the cell cycle and developmental transitions64,65. Cell cycle–resolved CUT&RUN confirmed this model, revealing a shift in the TgSNF2b/TgSNF2a ratio toward TgSNF2a at actively transcribed genes in each phase, and a return to TgSNF2b dominance at poised loci. Consistent with the role of TgSNF2b in developmental regulation, chromatin remodeling events involving SNF2b-associated factors have also been implicated in Plasmodium sexual development66. Specifically, disruption of PfARID resulted in a complete loss of male gametocyte formation, despite normal asexual blood-stage replication66,67. The preferential association of TgSNF2b with genes displaying brief expression windows was less pronounced for M-phase genes. This sustained promoter association may reflect a Toxoplasma-specific adaptation of the SWI/SNF-dependent mitotic bookmarking mechanism observed in mammalian systems, whereby the subunits SMARCE1 and SMARCB1 remain attached to the chromatin during mitosis to bookmark genes for reactivation while the ATP-dependent SWI/SNF helicase BRG1 is evicted68. However, in Toxoplasma, mitosis largely overlaps with DNA synthesis and cytokinesis and is accompanied by a single major chromatin remodeling event starting in late S phase and ending in late cytokinesis6. This results in a more continuous cell cycle compared to higher eukaryotes and may require TgSNF2a to remain attached to ensure continuous gene expression41. Consistent with this idea, our cell cycle–resolved CUT&RUN data revealed that several G1-phase genes exhibit strong TgSNF2a enrichment not only in G1 but also in M/C phases. This pattern is further supported by a study showing that components of both SWI/SNF complexes are differentially phosphorylated in response to TgCrk4 knockdown, a kinase essential for mitotic entry69. We correlated the distinct genome-wide binding patterns of TgSNF2a and TgSNF2b with published chromatin datasets. This analysis showed that TgSNF2a is enriched at active promoters and correlates with classical markers of euchromatin, such as H3K4me3, H4K31ac, and H3K14ac52. Accordingly, TgSNF2a occupancy is elevated at constitutively expressed genes but diminished at genes that are predominantly transcriptionally silent, reinforcing the notion that TgSNF2a associates with genes as they become transcriptionally active. By contrast, TgSNF2b colocalized with the repressive marks TgMORC and TgHDAC3, but also marks of activation like H4K31ac7,44,52. This pattern of chromatin association highlights a dual role for TgSNF2b in both gene repression and activation. Our regression analysis of variables associated with transcriptional changes following SWI/SNF perturbations revealed that chromatin context, rather than remodeler occupancy alone, best predicts transcriptional sensitivity. Genes marked by high levels of H4K31ac were more likely to be repressed following TgSNF2a, TgSNF2b, or TgSWI3 depletion. Conversely, genes upregulated upon TgSNF2b knockdown were enriched for H3K4me3 and TgSNF2b occupancy, and likely represent genes maintained in a poised state through a TgSNF2b-dependent mechanism. Genes upregulated upon TgSNF2a knockdown lacked strong associations with individual chromatin marks. This may indicate that the relevant chromatin features are either heterogeneous or that the correct chromatin context for TgSNF2a-dependent de-repression remains unidentified. A compelling candidate is the transcriptional repressor REST, whose DNA binding depends on the ATP-dependent SWI/SNF helicase BRG1 in mammalian systems70. While a clear REST ortholog has not been identified in Toxoplasma, the presence of structural homologs to key REST cofactors—such as RCOR1—suggests that a functionally analogous repressive complex may exist. Together, these findings support a model in which TgSNF2a and TgSNF2b function in a coordinated but non-redundant manner to regulate gene expression (Fig. 7E). TgSNF2b functions primarily to establish and maintain promoter accessibility, acting in concert with repressive factors such as TgMORC and TgHDAC3 to poise chromatin at developmentally regulated or cell cycle–dependent genes. TgSNF2a, by contrast, facilitates transcriptional reactivation at these accessible promoters, a function that becomes critical for genes with high temporal specificity, expressed during mitosis and cytokinesis. This division of labor ensures that promoter accessibility is maintained by TgSNF2b, while TgSNF2a reinstates or sustains transcription, ensuring coordinated gene expression and developmental timing. Methods Plasmids and primers Oligos were ordered from IDT. All cloning was performed with Q5 2 × master mix (NEB). All oligos used or generated in this study can be found in Supplementary Table 2. Parasite transfection T. gondii parasites were collected by centrifugation at 1000×g for 10 min and resuspended in Cytomix buffer (10 mM KPO4, 120 mM KCl, 0.15 mM CaCl2, 5 mM MgCl2, 25 mM HEPES, 2 mM EDTA, 2 mM ATP, and 5 mM glutathione). The cell suspension was mixed with transfection constructs to a final volume of 400 µl, and electroporation was performed using an ECM 830 Square Wave electroporator (BTX) with 4 mm cuvettes, applying two 1.7 kV pulses for 150 µs each at an interval of 100 ms. Strain generation All Myb domain–containing proteins as well as TgSNF2a and TgSNF2b were endogenously tagged using a high-throughput tagging (HiT) strategy24. Cutting units consisting of homology regions and sgRNA specific to each gene were purchased as gBlocks from IDT and integrated into the HiT vector backbone pGL015 (GenBank: OM640005) or pALH086 (GenBank ON312869) for TGGT1_237520 via Gibson assembly. Vectors were linearized using BsaI-HF V2 (NEB) and co-transfected with the Cas9 expression plasmid pSS014 (GenBank: OM640002) into Type I RH/TIR1/∆ku80/∆hxgprt or Type II ME49/TIR1/∆ku80/∆hxgprt parasites. Clones were selected with 3 µM pyrimethamine or 25 µg ml−1 mycophenolic acid and 50 µg ml−1 xanthine, isolated via limiting dilution and verified by PCR and immunofluorescence. TgSNF2a, TgSNF2b and TgTFIIIB were endogenously tagged with 2×ALFA48 using the HiT vector strategy by introducing gene-specific gBlocks into pDS033 empty HiT vector which is based on pGL015 (GenBank: OM640005) but possesses a XTEN-2×ALFA tagging unit. The vector was linearized with BsaI-HF V2 (NEB) and co-transfected with the Cas9 expression plasmid pSS014 (GenBank: OM640002) into Type I RH/∆ku80/∆hxgprt parasites. Clones were selected with 3 µM pyrimethamine. Thymidine synchronization strains containing a chloramphenicol acetyl transferase-herpes simplex virus thymidine kinase (CAT-HSVTK) fusion protein ectopically expressed from empty region 1 (ER1, Chr. IV)71 were generated by co-transfecting pBM041 (GenBank: MN019116) with a CAT-HSVTK amplicon containing the TgTub1 5’UTR as well as the TgDHFR 3’UTR sequences and ER1 homology arms generated from the original CAT-HSVTK plasmid generously provided by Michael W. White51. TgSNF2a and TgSNF2b were endogenously tagged with 2×ALFA48 using the HiT vector strategy by introducing gene-specific gBlocks into pDS126, which is based on pDS033 but possesses hxgprt instead of DHFR. Clones were selected with 40 µM chloramphenicol or 25 µg ml−1 mycophenolic acid and 50 µg ml−1 xanthine. Cell cycle reporter strains expressing IMC1 endogenously tagged with mScarlet-I72 as well as H2B.Z tagged with BFP73 ectopically expressed from empty region 1 (ER1, Chr. IV)71 were generated by co-transfecting the following constructs into Type I RH/TIR1/∆ku80/∆hxgprt: For IMC1-mScarlet-I, a sgRNA targeting the IMC1 locus was introduced into the Cas9 expression plasmid pSS013 (GenBank: OM640003) which was co-transfected with an XTEN-mScarlet-I amplicon generated from pDS131 empty HiT vector which is based on pALH19374 but possesses mScarlet-I derived from Addgene plasmid #125138 instead of mNG. The amplicon was generated using primers containing IMC1 homology arms. A copy of H2B.Z-BFP was introduced to ER1 by co-transfection of pBM041 (GenBank: MN019116) with an H2B.Z-BFP amplicon containing the TgTub1 5’UTR as well as the TgDHFR 3’UTR sequences and ER1 homology arms generated from pDS118, which was derived from pH2B-YFP25 by replacing the YFP sequence with the sequence of BFP derived from Addgene plasmid #171090. Clones were selected by flow cytometric sorting of double-positive parasites expressing both mScarlet-I and BFP. TgTFIIIB endogenously tagged at the C-terminus using the HiT strategy was additionally tagged at the N-terminus with a 2×ALFA epitope tag by co-transfecting pSS013 containing a sgRNA targeting TgTFIIIB at the N-terminus, as well as a 2×ALFA-XTEN amplicon generated with primers containing TgTFIIIB homology arms. Clones were selected by limiting dilution. A complete list of strains with their respective genotype and origin is provided in Supplementary Table 3. Cell culture Toxoplasma gondii parasites and human foreskin fibroblasts (HFFs, ATCC SRC-1041) were cultured in Dulbecco’s Modified Eagle Medium (DMEM, GIBCO 11965118) supplemented 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts) and 3% inactivated calf serum (ICS) for RH parasites or 3% inactivated fetal bovine serum (IFS) for ME49 parasites unless specified otherwise. Tachyzoites and HFFs were grown at 37 °C and 5% CO2. Stress conditions were induced using alkaline stress medium consisting of RPMI 1640 (Sigma) supplemented with 1% IFS, 10 μg/ml gentamicin, and buffered with 50 mM HEPES, adjusted to pH 8.1 using 10 N NaOH. HFFs and parasites were routinely tested for mycoplasma using the ATCC Universal Mycoplasma Detection Kit (30-1012 K). Cell cycle synchronization Parasites containing CAT-HSVTK were synchronized with excess thymidine based on Radke and White51. Briefly, parasites were grown for 24 h in DMEM (GIBCO 11965118) supplemented with 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts) and 10% IFS. At 24 h post-infection, the media was exchanged for fresh media containing 40 µM thymidine (Sigma-Aldrich) to synchronize the population. After 6 or 7 h of growth in thymidine media, early S or mid S phase parasites were harvested, respectively. For G1a/b and M/C phase parasites, the thymidine-containing media was aspirated, followed by two washes with pre-warmed PBS and the addition of fresh DMEM (GIBCO 11965118) supplemented with 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts) and 10% IFS. After 1 or 5 h of post-wash growth, M/C or G1a/b phase parasites were harvested. Immunohistochemistry HFFs were grown on coverslips for 2–6 days in DMEM (GIBCO 11965118) supplemented with 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts) and 10% IFS to form a confluent monolayer before infection with Toxoplasma. Intracellular parasites were fixed for 20 min in ice-cold 4% formaldehyde in PBS, permeabilized with 1% Triton X-100 in PBS for 8 min, and blocked for 10 min in PBS with 5% normal goat serum and 5% IFS. Primary and secondary antibody incubations were performed for 1 h each, with three intermediate PBS washes after each step. Before mounting, coverslips were washed by a brief immersion in dH₂O and mounted on Prolong Diamond (Thermo Fisher Scientific) at 37 °C for 30 min or overnight at room temperature. GAP45 was detected using the rabbit anti-GAP45 antibody (1:5000 dilution), which was a gift from Dominique Soldati (University of Geneva)75. Mouse anti-mNG (ChromoTek) was used at 1:1000. Mouse anti-IMC1 was a gift from Gary Ward (University of Vermont)76 and used at 1:2000. Guinea pig anti-CDPK1 (Covance) was used at 1:10,000. Rabbit anti-Centrin1 (Kernfast) was used at 1:500. The ALFA-tag was detected using either FluoTag-X2 anti-ALFA-647 (NanoTag Biotechnologies) at 1:1000 or rabbit anti-ALFA (NanoTag Biotechnologies) at 1:1000. Cyst walls were stained with Dolichos biflorus lectin (DBL-488 or DBL-575; 1:500 or 1:200, VectorLabs). Secondary antibodies conjugated to Alexa Fluor 488, 594 or 647 (Thermo Fisher Scientific) were used at 1:1000, and DNA was counterstained with Hoechst 33258 (1:20,000, Life Technologies). Images were acquired using an Eclipse Ti-E epifluorescence microscope or a Yokogawa Spinning Disk Field Scanning Confocal microscope and processed in FIJI (v2.14.0/1.54 f)77. The quantifications of immunohistochemistry data generated in this study are provided in the Source Data file. Differentiation assay Confluent HFF monolayers grown on coverslips were infected with parasites and cultured in DMEM (GIBCO 11965118) supplemented with 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts) and 3% IFS. At 24 h post-infection, the media was exchanged for fresh media containing either 500 µM auxin (indole-3-acetic acid) or vehicle (PBS). After 4 h of auxin or vehicle treatment, parasites were shifted to stress conditions (alkaline stress medium) supplemented with 500 µM auxin or vehicle. Stressed parasites were cultured at 37 °C and ambient CO2 for 48 h before cells were fixed and stained as described under “Immunohistochemistry”. Parasites were probed with antibodies against mNG as well as DBL-575 and Hoechst dye to label the parasite’s cyst wall or DNA, respectively. DBL staining was quantified for ≥100 vacuoles per replicate. The differentiation assay data generated in this study are provided in the Source Data file. Replication assay Confluent HFF monolayers grown on coverslips were infected with parasites and cultured in DMEM (GIBCO 11965118) supplemented with 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts), 3 % ICS and 50 µM auxin or vehicle (PBS), respectively. At 24 h post-infection, cells were fixed and stained as described under “Immunohistochemistry”. Parasites were probed with antibodies against GAP45 and mNG as well as Hoechst dye to label DNA. Parasites per vacuole were quantified for ≥100 vacuoles per replicate. The replication assay data generated in this study are provided in the Source Data file. Video microscopy Confluent HFF monolayers were grown in glass-bottom imaging dishes (VWR International) and infected with parasites ~10 h before imaging and grown in FluoroBrite™ DMEM (Thermo Fisher Scientific) supplemented with 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts), 10% IFS and 50 µM auxin or vehicle (PBS), respectively. Parasites were imaged at 37 °C and 5 % CO2 for 24 h at 1 frame every 20 min using a Dual-Fusion BT camera or at 1 frame every 30 min using a Fusion BT camera. Photobleaching was evaluated by imaging parasites over a short interval while matching the total frame count used in the 24-h acquisition. Videos were cropped and annotated using NIS elements (Nikon) and FIJI (v2.14.0/1.54 f)77. Flow cytometry analysis Parasite DNA content following TgSNF2a and TgSNF2b knockdown was evaluated by flow cytometry using a modified propidium iodide (PI, Sigma-Aldrich) staining protocol69. HFF monolayers were infected with parasites and cultured for 24 h at 37 °C with 5% CO₂ before they were treated with 50 µM auxin or vehicle (PBS) for 12, 24, or 48 h. Following treatment, monolayers were aspirated, rinsed with PBS, and parasites were mechanically released by scraping and passage through a 27-gauge needle, followed by filtration through a 5-μm membrane. Parasite suspensions were centrifuged at 1500×g for 10 min and fixed by resuspension of the pellet in 70% ice-cold ethanol. Samples were stored at –20 °C until flow cytometry analysis. On the day of analysis, parasites were pelleted at 4 °C at 800×g for 5 min, the supernatant was aspirated, and pellets were resuspended in 1 ml ice-cold PBS. PI was added to a final concentration of 0.1 mg/ml, along with RNaseA/T1 mix (Thermo Fisher Scientific) to 0.08 mg/ml. Samples were incubated for 30 min in the dark at room temperature. DNA content was measured by flow cytometry using a MACSQuant VYB Flow Cytometer (Miltenyi Biotec). Flow cytometry data were analyzed using FlowJo (v10.10.0). The quantification of the DNA content data generated in this study are provided in the Source Data file. Plaque assay Plaque assays were performed by infecting a six-well plate containing a confluent monolayer of HFFs with 500 parasites per well. Parasites were cultured in DMEM (GIBCO 11965118) supplemented 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts), 10% IFS and 50 µM auxin or vehicle (PBS), respectively. Cultures were maintained undisturbed for 10 days, then washed with PBS and fixed in 100% ethanol for 10 min at room temperature. Plates were air-dried, stained with crystal violet (12.5 g crystal violet, 125 ml 100% ethanol, and 500 ml 1% ammonium oxalate) for 1 h, washed three times with PBS and once with water, and air-dried prior to imaging. RNA-sequencing RNA-sequencing library preparation Parasites were seeded onto confluent HFF monolayers in 10 cm dishes and allowed to replicate for 24 h in DMEM (GIBCO 11965118) supplemented with 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts), 3% ICS. Thereafter, the media was replaced with DMEM (GIBCO 11965118) supplemented with 10 μg/ml gentamicin (Thermo Fisher Scientific), 2 mM L-glutamine (Gemini Bioproducts), 3% ICS and 50 µM auxin or vehicle (PBS), respectively. Parasites were cultured for an additional 6, 12, or 24 h before harvest. Monolayers were aspirated, rinsed with PBS, and parasites mechanically released by scraping and passage through a 27-gauge needle, followed by filtration through a 5-μm membrane. Parasite suspensions were centrifuged at 1500×g for 5 min, and RNA was extracted from parasite pellets by double extraction using TRIzol (Invitrogen). Therefore, the pellet was resuspended in 800 µl TRIzol Reagent (Invitrogen), vortexed, and split into two different tubes to retain one technical replicate. Phase separation was performed in phase lock heavy tubes (Quantabio) by the addition of 100 μl chloroform:isoamyl alcohol (24:1, Sigma-Aldrich), vigorous vortexing (1 min), and centrifugation at 16,000×g for 5 min at 4 °C. The aqueous phase was isolated, and TRIzol was extracted again using 400 µl TRIzol. The resulting aqueous phase was supplemented with 1 μl Glycogen (20 mg/ml, Thermo Fisher Scientific) and 500 μl isopropanol, and RNA was precipitated at room temperature for 12 min. Pellets were recovered by centrifugation (12,000×g, 12 min) at 4 °C, washed twice with 70% ethanol, air-dried for 10 min at room temperature, and resuspended in 20 μl RNase-free dH2O. RNA quality was assessed using a Bioanalyzer (Agilent). RNA-seq libraries were prepared with the Rapid Directional RNA-Seq Kit 2.0 (NEXTFLEX) using Poly(A) Beads 2.0 (NEXTFLEX). Paired-end sequencing (50/50 or 100/100) was performed on an Illumina NovaSeq SP platform. RNA-sequencing data processing and analysis Reads were trimmed using Trim Galore (v0.6.7) and aligned to the Toxoplasma gondii ME49 reference genome (ToxoDB v68) with STAR (v2.7.1a) using the --quantMode GeneCounts option and the parameters --outFilterMismatchNoverLmax 0.04 and --outFilterMatchNminOverLread 0.4 to control alignment stringency. Differential-expression analysis was performed using the DESeq2 R package (v. 1.40.2)78. Genes were classified as up- or down regulated using the Benjamini–Hochberg adjusted p value. Genes with an adjusted p value <0.005 and a fold change >2 were defined as upregulated, while genes with an adjusted p value <0.005 and a fold change <–2 were defined as downregulated. To minimize noise from low-abundance transcripts, only genes with a normalized mean read count greater than 17.3 or 37.9 were considered differentially regulated; genes below this threshold were excluded from subsequent analyses based on their expression distribution. Replicate 1 of the TGME49_275480 knockdown was excluded from further analysis due to underrepresentation in the sequencing libraries (Supplementary Fig. 2A). For the implementation of Gini scores into the differential-expression analysis, Gini scores were calculated from published transcriptome data obtained from synchronized parasites43. Spline smoothed normalized sample values for each gene were downloaded from ToxoDB (www.toxodb.org) at 13 different time points after release from thymidine synchronization. After removing genes with missing values, Gini coefficients were computed across time points for each gene using the ineq R package (v0.2.13). The resulting Gini score reflects expression inequality across the cell cycle, with higher values indicating more stage-restricted expression and lower values indicating uniform expression across time points. The sequencing data generated in this study have been deposited in the Gene Expression Omnibus79 database under the accession code GSE303610. It should be noted that in previous versions of this manuscript, TgSNF2a was designated as TgBRG1 and TgSNF2b as TgBRM. Immunoprecipitation Anti-mNG immunoprecipitation Anti-mNG-Trap magnetic agarose resin (ChromoTek) was used for anti-ALFA immunoprecipitations, and anti-ALFA Selector PE magnetic agarose resin (NanoTag Biotechnologies) was used. Beads were pre-eluted using 50 mM HEPES-NaOH (pH 7.5), 0.4% IGEPAL and 2 mM Dithiothreitol (DTT) in dH2O. Nuclear lysates of ~2.5 × 107 intracellular parasites were generated using the NE-PER Nuclear and Cytoplasmic Extraction Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol. The nuclear fraction was added to the pre-eluted beads and incubated at room temperature for 30 min at rotation, followed by three washes with 50 mM HEPES-NaOH (pH 7.5), 0.04% IGEPAL and 2 mM DTT in dH2O. Elution of bound proteins from beads was performed by incubation of beads in elution buffer (5% SDS, 40 mM chloroacetamide, 10 mM Tris(2-carboxyethyl)phosphine) and 100 mM Triethylammonium bicarbonate (TEAB) in dH2O) for 10 min at 70 °C and slow mixing. The elution was transferred to a new tube, and the elution was repeated. Elutions were pooled, and the standard S-trap micro workup protocol (Profiti) was used to prepare samples for LC-MS/MS collection. In brief, samples were acidified with 2.5 μl 27.5% phosphoric acid and precipitated by adding 165 μl S-trap binding buffer (100 mM TEAB in 90% ice-cold methanol). The total volume was applied to S-trap Micro columns, and the precipitated proteins were separated from suspension via centrifugation for 1 min at 4000×g. Detergents and salts were removed by washing the S-trap columns six times with 150 μl S-trap binding buffer. Proteins caught by the S-trap fiber glass filters were enzymatically digested by adding 1 μg trypsin/LysC mix dissolved in 30 μl 100 mM TEAB and incubation overnight at 37 °C in a humidified incubator. Tryptic peptides were collected by centrifuging for 1 min at 4000×g followed by sequential washes with 40 μl of 50 mM TEAB, pH 8.5, 40 μl 0.2% formic acid in water, and 40 μl 50% acetonitrile in water. The resulting peptides were pooled, lyophilized, and dissolved in 12 μl 0.2% formic acid in MS-grade water. LC-MS/MS Data were generated using an Easy-nLC 1200 system coupled with an Orbitrap Exploris 480 mass spectrometer, both from Thermo Fisher Scientific (Waltham, MA, USA). The setup included a FAIMS Pro Interface and an Easy Spray ESI source. NanoLC separation utilized an Acclaim PepMap trap column (75 μm × 2 cm) (Thermo Fisher Scientific) combined with an Aurora Ultimate TS25 column (75 µm × 25 cm, 120 Å) from IonOpticks (Fitzroy, VIC, AUS). A 5 µl volume of peptide extracts was injected. Peptide separation was conducted with a mobile phase consisting of 0.1% formic acid in water (solution A) and 0.1% formic acid in 80 % acetonitrile (solution B), flowing at 400 nl/min, while the column temperature was held constant at 50 °C. Peptides were separated on a gradient of 3–25% solution B for 68 min, 25–40% solution B for 22 min, 40–95% solution B for 3 min and 95% solution B for 7 min. Using the MS in positive mode, the ion source temperature was set to 270 °C, and ionized peptides were passed through the FAIMS Pro at −45 and −65 V. Mass spectra were collected in MS1 mode with a resolution of 60,000, spanning the mass range of 350–1400 m/z using custom automatic gain control (AGC = 300) settings and an injection time of 25 ms. For MS2 data acquisition, the MS was operated in DDA mode with dynamic exclusion and a resolution of 15,000. The top precursors with charge states 2–6 were selected for MS2 analysis using a dynamic exclusion window of 45 s. The following parameters were used for MS2 acquisition: automatic maximum injection time, standard AGC target, isolation window of 1.3 m/z, 30% normalized collision energy, and an intensity threshold of 5 × 103. LC-MS/MS data analysis Analysis was performed using Proteome Discoverer (v2.4.1.15) for peptide identification and quantification. Downstream data processing was conducted using R, incorporating the packages dplyr (v1.1.4) and readxl (v1.4.3). Specifically interacting proteins were defined as those identified with >2 unique peptides in n = 2 biological replicates for TgSWI3 and n = 3 biological replicates for TgSNF2a and TgSNF2b, showing an enrichment over the untagged control of >3-fold for TgSWI3 and TgSNF2b or >2-fold for TgSNF2a, and meeting a Benjamini–Hocherg-adjusted two-way ANOVA p value threshold of <0.05. The mass spectrometry proteomics data generated in this study were deposited in the ProteomeXchange Consortium database via the PRIDE partner repository80 under the accession code PXD066820. It should be noted that in previous versions of this manuscript, TgSNF2a was designated as TgBRG1 and TgSNF2b as TgBRM. CUT&RUN NbALFA-MNase generation The His-tagged NbALFA-MNase fusion protein construct was generated from the original pAG-MNase fusion protein construct generously provided by Mary Gehring47. The original vector was digested with AfeI (NEB) and mluI-HF (NEB) to replace the endogenous pAG sequence with the NbALFA sequence derived from the original ALFA-tag publication48. The DNA plasmid was transformed into chemically competent JM101 cells (Agilent) according to the manufacturer’s instructions and cultured in NZYM medium supplemented with 50 μg/ml kanamycin. At an OD₆₀₀ of ~0.6, NbALFA-MNase expression was induced with 2 mM isopropyl-β-D-1-thiogalactopyranoside (IPTG; Thermo Fisher Scientific) for 2 h at 37 °C. Cells were harvested by centrifugation (3500×g, 20 min), washed in ice-cold PBS, and lysed by resuspension in 10 ml lysis buffer (10 mM Tris-HCl, pH 7.5, 300 mM NaCl, 10 mM imidazole, 5 mM β-mercaptoethanol) containing 10 μg/ml lysozyme (Sigma-Aldrich). After 15 min at 30 °C, cells were lysed using a microfluidizer at 4 °C. Lysates were clarified by centrifugation (12,000×g, 20 min, 4 °C) and loaded onto a Ni-NTA agarose column (Qiagen). Binding was performed for 1 h at 4 °C on a rotator, and columns were washed with wash buffer (10 mM Tris-HCl, pH 7.5, 300 mM NaCl, 20 mM imidazole, 0.03% Zwittergent 3–10). The NbALFA-MNase fusion protein was eluted with elution buffer (10 mM Tris-HCl, pH 7.5, 300 mM NaCl, and 250 mM imidazole) and dialyzed twice at 4 °C (2 h and overnight) against dialysis buffer (10 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, and 1 mM PMSF) using Slide-A-Lyzer cassettes (Thermo Fisher Scientific). Purified protein was supplemented with glycerol to 50% final concentration and stored at −80 °C. The quality of the NbALFA-MNase purification was evaluated by Coomassie Brilliant Blue staining, and enzymatic activity was assessed by the digestion of high-molecular-weight lambda DNA (NEB). ALFA-CUT&RUN Intracellular parasites were harvested by mechanical disruption through a 27-gauge needle and filtration through a 5-μm membrane. A total of ~5 × 10⁷ parasites were pelleted (1500×g, 10 min) and resuspended in wash buffer (20 mM HEPES-NaOH, pH 7.5, 150 mM NaCl, 0.5 mM Spermidine, and 1x EDTA-free Protease Inhibitor Cocktail) and attached to 100 μl BioMag Plus Concanavalin A beads (Polysciences) activated in bead activation buffer (20 mM HEPES-KOH, pH 7.9, 10 mM KCl, 0.1 mM CaCl2, and 0.1 mM MnCl2). Parasites were bound to the beads by incubation for 15 min at room temperature and rotation. Bead-bound parasites were blocked with wash buffer supplemented with 2 mM EDTA for 5 min at room temperature, followed by binding of 1 µg NbALFA-MNase per sample in ice-cold wash buffer supplemented with 0.025% Digitonin (Invitrogen) for 1 h at 4 °C. After NbALFA-MNase binding, bead-bound parasites were washed three times with ice-cold wash buffer supplemented with 0.025 % Digitonin for 5 min at 4 °C. The cutting reaction was started by the addition of ice-cold wash buffer supplemented with 0.025% Digitonin and 2 mM CaCl2, and incubation for 60 min on ice. The reaction was stopped by the addition of 4×STOP buffer (400 mM NaCl, 40 mM EDTA, 8 mM EGTA, 100 µg/ml RNaseA, 80 µg/ml Glycogen, and 0.025% Digitonin) to 1× and incubation at 37 °C for 20 min. The DNA was extracted using a modified organic extraction and ethanol precipitation protocol3. CUT&RUN library preparation CUT&RUN libraries were prepared using the NEBNext Ultra II DNA Library Prep Kit for Illumina (NEB) and NEBNext Multiplex Oligos for Illumina (NEB) according to the manufacturer’s instructions with 11–12 PCR amplification cycles. Size selection and clean-up steps were performed with AMPure XP beads (Beckman Coulter). Library quality and fragment size distribution was assessed using a Bioanalyzer (Agilent). Paired-end sequencing (75/75) was performed on an Illumina MiSeq v3 platform, generating ~6 million reads per sample. CUT&RUN data processing and analysis Paired-end reads were trimmed using Trim Galore (v0.6.7) with a Phred quality cutoff of 20 and discarding reads shorter than 30 bp. Trimmed reads were aligned to the Toxoplasma gondii ME49 reference genome (ToxoDB v68) using Bowtie2 (v2.4.2) with the parameters: --end-to-end -D 15 -R 2 -N 1 -L 22 -i S,1,0.50 --no-mixed --no-discordant --phred33 -I 10 -X 700. For asynchronous parasite population CUT&RUN experiments, peak calling was performed with MACS2 (v2.2.7.1) using the following options: -f BAMPE -g 65.67e6 -B --max-gap 0 -p 0.05 --keep-dup all --nomodel --call-summits. Peak refinement and annotation was carried out using R, incorporating the packages utils (v4.3.1), tidyverse (v2.0.0), tidyr (v1.3.1), readxl (v1.4.3), and dplyr (v1.1.4). Peaks containing two summits separated by >1000 bp with a pileup difference <40% were split into distinct peaks. Reproducibility was assessed using IDR2D (v1.14.0). Peaks reproducible at p < 0.05 and those also identified in the TgTFIIIB control (fold enrichment >1.5 and p < 0.05) were assigned to gene IDs based on proximity to the nearest TSS, accounting for gene orientation, using plyr (v1.8.9), DescTools (v0.99.58), ape (v5.8.1), openxlsx (v4.2.7.1), GenomicRanges (v1.52.1), and stringr (v1.5.1). Peaks equidistant to two genes (<1.5-fold difference) were assigned to both. Differential-binding analysis was performed using DiffBind (v3.10.1) with DESeq2 as the statistical framework and a minimum quality control threshold of 15. Regions were defined as windows centered on the peak summit with a total length corresponding to two mean fragment sizes. Relative enrichment of TgSWI3, TgSNF2a, or TgSNF2b within ±500 bp of annotated TSSs was assessed by computing kernel density estimates of peak-to-TSS distances. Density curves were generated using ggplot2 with default parameters. Differential enrichment between TgSNF2b and TgSNF2a was defined using the Benjamini–Hochberg adjusted p value combined with a fold change cutoff: peaks with adjusted p value <0.05 and fold change >2 were classified as TgSNF2b enriched, while those with adjusted p value <0.05 and fold change <–2 were classified as TgSNF2a enriched. For RNA-seq integration, if multiple peaks were associated with a gene, only the closest peak was retained. The sequencing data generated in this study have been deposited in the Gene Expression Omnibus79 database under accession code GSE303606. It should be noted that in previous versions of this manuscript, TgSNF2a was designated as TgBRG1 and TgSNF2b as TgBRM. For cell cycle–resolved CUT&RUN experiments, consensus peak regions defined in asynchronous parasites were used for differential-binding analysis performed in R using DiffBind (v3.10.1) with the DESeq2 method for library size normalization and a minimum quality control threshold of 15. These normalized counts were used to quantify TgSNF2a and TgSNF2b occupancy across G1a/b, early S, mid S, and M/C, and to assess temporal changes at high-confidence peak-gene pairs. The sequencing data generated in this study have been deposited in the Gene Expression Omnibus79 database under the accession code GSE313318. It should be noted that in previous versions of this manuscript, TgSNF2a was designated as TgBRG1 and TgSNF2b as TgBRM. Phylogenetic analysis Phylogenetic trees were generated based on the identification of Myb domains (SM00717) in T. gondii18. Proteins containing a DEXDc (SM00487) and a HELICc (SM00490) domain were identified in a broadly representative group of apicomplexan species, including T. gondii, Eimeria tenella, Plasmodium falciparum, Plasmodium berghei, Theileria annulata, Babesia microti, Cryptosporidium parvum, and Gregarina niphandrodes and the related species Chromera velia, through a homology-based search with SMART 9.0 (e-value <0.05). The initial set of 604 DEXDc+HELICc domain–containing proteins was further enhanced by 40 additional DEXDc+HELICc domain–containing proteins using BLASTP81,82 against all analyzed species and filtering by domain length of the initial protein set identified by SMART (min. length = 0.5x DEXDc: 81 AA, 0.5x HELICc: 35 AA; max. length = 1.5x DEXDc: 1848 AA, 1.5x HELICc: 398 AA). Saccharomyces cerevisiae, Drosophila melanogaster, and Homo sapiens were included as references in certain comparisons. Reciprocal BLASTP using the newly identified DEXDc+HELICc domain–containing proteins resulted in the identification of three additional DEXDc+HELICc domain–containing proteins. An additional reciprocal BLASTP search did not result in any additional hits, resulting in 647 DEXDc+HELICc domain-containing of which 85 were Toxoplasma. We performed phylogenetic analysis on the 17 Toxoplasma proteins that showed the highest sequence similarity to TgSNF2a and TgSNF2b (e-value cutoff: 0.01). To generate a maximum likelihood (ML) phylogenetic tree, we concatenated all DEXDc and HELICc domains for each protein and aligned them using the MUSCLE implementation of MEGA11 (v11.0.13). The best model to construct the ML tree was calculated using ModelTest-NG (v0.2.0). The best-scoring model available in MEGA11 (LG + G4 + F) was used to construct the ML tree using MEGA11 with 1000 bootstrap replicates. The closest relative to TgSNF2a and TgSNF2b with an e-value >0.01 (TGME49_231800) was used as an outgroup. Homology-based searches Whole-proteome FASTA files were obtained from EuPathDB (v68) and NCBI RefSeq databases (release 225). Protein datasets were formatted for local sequence similarity searches using makeblastdb from the BLAST+ suite (v2.15.0)81,82. Homology searches were performed locally using BLASTP from the BLAST+ suite, with each protein used as a query. Searches were conducted with parameters optimized for sensitivity, using a word size of 5 and a maximum of 10,000 alignments. For each species, the best hit corresponding to the lowest e-value was retained for subsequent analysis to assess evolutionary conservation and lineage-specific retention. Immunoblotting TgSNF2a-mAID/TgSNF2b-2×ALFA and TgSNF2a-2×ALFA/TgSNF2b-mAID parasites were grown in standard media for 24 h, followed by treatment with either vehicle or 50 µM auxin for the duration of the indicated time points (0, 3, 6, 12, and 24 h). 2×ALFA-TGGT1_203950-mNG-V5-mAID-Ty parasites were grown in standard media for 24 h. About 106 intracellular parasites were scraped, passed through a 27-gauge needle, filtered through a 5 µm filter and pelleted. The parasite pellets were resuspended in Laemmli buffer (2% SDS, 10% glycerol, 60 mM Tris-HCl, pH 6.8, 0.01% bromophenol blue, and 1% β-mercaptoethanol) and run on a 4–15% SDS-PAGE gel (Bio-Rad). Proteins were transferred onto a nitrocellulose membrane using the Trans-Blot® Turbo™ Transfer System and the high or mixed molecular weight protocol (high: 1.3 A and up to 25 V for 10 min; mixed: 1.3 A and up to 25 V for 7 min). Membrane blocking and subsequent antibody probing was performed with rocking for 1 h at room temperature in TBST (Tris-buffered saline with 0.1% Tween-20) containing 5% milk. Between primary and secondary antibody probing and before imaging, membranes were washed three times with TBST. Horseradish peroxidase-conjugated secondary antibodies were imaged using the Radiance Plus Femtogram HRP substrate (Azure Biosystems) and a ChemiDoc imaging system (Bio-Rad). Uncropped and unprocessed immunoblot images are provided in the Source Data file. Chromatin feature extraction and analysis The wig and bigwig data used in this study are available in the Gene Expression Omnibus79 database and were obtained from ref. 52 (H3K9me3: GSM2612412, H4K20me3: GSM2612413, H3K4me1: GSM2612414, H3K4me3: GSM2612415, H3K14ac: GSM2612416, H4K31me1: GSM2612417, H4K31ac: GSM2612418) as well as ref. 44 (H3.3: GSM2795864, H2AX: GSM2795867, H2AZ: GSM2795868, H2BV: GSM2795870) and ref. 7 (MORC: GSM4040359, HDAC3: GSM4040363). Wig and bigwig files were analyzed using R, relying on the R packages wig (v0.1.0), GenomicRanges (v1.52.1), ape (v5.8.1), scales (v1.3.0), and rtracklayer (v1.60.1). Signal intensities from wig and bigwig files were quantified over gene bodies as annotated in the T. gondii ME49 reference genome (ToxoDB v68) and TgSNF2a/TgSNF2b consensus peaks and averaged per region. Gene body and SWI/SNF consensus promoter signals were log2 transformed, mean-centered and filtered to retain chromatin features of interest, followed by calculation of Pearson correlations to assess relationships between chromatin marks and remodeler occupancy. Chromatin features predictive of transcriptional changes following TgSNF2a, TgSNF2b, or TgSWI3 knockdown were identified using R, relying on the R package dplyr (v1.1.4). We filtered genes for sufficient expression (normalized counts >37.93) and retained those containing a SWI/SNF consensus promoter signal. We selected both promoter-associated and gene body chromatin features, including histone modifications and remodeler occupancy, as predictors. The signal of each predictor from wig and bigwig files were quantified over gene bodies as annotated in the T. gondii ME49 reference genome (ToxoDB v68) and TgSNF2a/TgSNF2b consensus peaks, averaged per region and z-score normalized. Baseline RNA expression levels were included as an additional covariate in all regression models. For each knockdown, a weighted multiple linear regression model was then fitted, using the respective log2 normalized fold changes as the response and −log10(adjusted p values) as weights. Model coefficients were extracted and visualized with confidence intervals to assess the direction and strength of each feature’s contribution. Data analysis and visualization All data analyses were performed using RStudio (v2023.06.0 + 421) and R (v4.3.1). Data manipulation and visualization were conducted using standard R packages, including ggplot2 (v3.5.2), ggbeeswarm (v0.7.2), ggpubr (v0.6.0), ggradar (v0.2), VennDiagram (v1.7.3), ComplexHeatmap (v2.16.0), scales (v1.3.0), RColorBrewer (v1.1.3) and viridis (v0.6.5) for plotting along with utils (v4.3.1), tidyverse (v2.0.0), tidyr (v1.3.1), readxl (v1.4.3), and dplyr (v1.1.4), tibble (v3.2.1) ineq (v0.2.13), purrr (v1.0.4) and reshape2 (v1.4.4) for data wrangling and visualization. Gene set enrichment analysis (GSEA) of cell cycle–associated gene sets was conducted using the R package fgsea (v1.26.0) with p values calculated based on an adaptive multi-level split Monte-Carlo scheme83. Fluorescence intensity from live-cell video microscopy was quantified using CellProfiler (v4.2.8)84. Fold change data for T. gondii sexual-stage transcriptomes were obtained from ToxoDB (www.toxodb.org) using the fold change option without applying any significance or fold change thresholds. Fold change data for T. gondii extracellular transcriptomes were obtained from ref. 85. The chromatin profiling sequencing reads for TgMORC used in this study are available from ref. 7 in the Gene Expression Omnibus79 database under accession code GSE136060. The analysis pipelines supporting the current study are available from the corresponding author upon request. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Supplementary information Supplementary Information Description Of Additional Supplementary File Supplementary Data 1 Supplementary Data 2 Supplementary Data 3 Supplementary Data 4 Supplementary Data 5 Supplementary Data 6 Supplementary Data 7 Supplementary Data 8 Supplementary Data 9 Supplementary Data 10 Supplementary Movie 1 Supplementary Movie 2 Supplementary Movie 3 Reporting summary Transparent Peer Review file Supplementary Information Description Of Additional Supplementary File Supplementary Data 1 Supplementary Data 2 Supplementary Data 3 Supplementary Data 4 Supplementary Data 5 Supplementary Data 6 Supplementary Data 7 Supplementary Data 8 Supplementary Data 9 Supplementary Data 10 Supplementary Movie 1 Supplementary Movie 2 Supplementary Movie 3 Reporting summary Transparent Peer Review file Source data Source data Source data