H2B.V demarcates T. cruzi divergent switch strand regions, some tDNA loci and genome compartments

H2B.V is present as a single-copy gene at chromosome 27 in the T. cruzi CL Brener genome assembly. Thus, to evaluate the genomic location of H2B.V, we generated parasites expressing tagged H2B.V at the C-terminus with 3 x myc peptides by CRISPR-Cas9 methodology (S1A Fig). The genomic edition of the H2B.V locus was confirmed by western blotting and immunofluorescence assay (S1C and S1G Fig). Allele-specific PCRs indicate that both alleles were edited (S1B Fig). The genome edition neither affects parasite growth nor the metacyclogenesis processes (S1D and S1E Fig). In addition, H2B.V transcript abundance in epimastigotes and TCTs do not differ from wild-type cells or parasites expressing Cas9 (S1F Fig). Together, these data indicated that the genomic edition at H2B.V locus confers no significant phenotypic changes.

To identify the genomic regions associated with nucleosomes harboring H2B.V, the chromatin of myc-tagged epimastigotes and TCTs was digested with micrococcal nuclease (MNase) and sonicated followed by chromatin immunoprecipitation (ChIP) using anti-myc and anti-histone H3 antibodies (as a control for the nucleosome distribution along genome). Furthermore, the input for all ChIP samples and an additional control sample of an untagged epimastigote cell line (immunoprecipitated with an anti-myc antibody) were also analyzed. To improve the data analysis and biological conclusions, the reads were mapped against two T. cruzi genome assemblies (the Esmeraldo-like haplotype of the CL Brener strain and against the TCC strain), and the results obtained for each were compared throughout this study. The first corresponds to the strain used for ChIP experiments, the genome of which was sequenced by second-generation sequencing strategies [18], and the second harbors 99.7% identity to the CL Brener genome and was recently sequenced by PacBio into long reads [21]. By mapping the reads to these two assemblies, we could evaluate whether any bias related to genome sequencing and assembly quality could interfere with the interpretation of our results. Approximately 2 million reads per replicate per ChIP-seq experiment were sequenced, generating approximately 85% and 90% overall alignment (S2A Fig) in the CL Brener and TCC assembly, respectively.

Visual inspection of H2B.V enrichment (Fig 1A) indicated that this variant was deposited along the edges of dSSRs, including, in some cases, the first few coding DNA sequences (CDSs) from the polycistronic regions (Figs 1A and S3A). In accordance, the summary plot of the normalized ratio of H2B.V (ChIP/input) enrichment clearly indicated that H2B.V was preferentially enriched in the upstream region of polycistrons (Fig 1B) in both the CL Brener (Fig 1B) and TCC assembly (S3B Fig). Through H2B.V peak enrichment data, we observed that H2B.V was deposited over large regions of approximately 5 kb in width (Figs 1C and S2C). dSSRs, but not cSSRs, were demarcated by a peak at each edge, and no specific histone H3 enrichment was found at these regions (Fig 1C).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. H2B.V is enriched at dSSRs of epimastigote forms.

A. IGV snapshot of whole chr 6S of the CL Brener Esmeraldo-like haplotype using the wig files from the ChIP H2B.V/input and ChIP H3/input ratios obtained by COVERnant v.0.3.2. Note that H2B.V, but not H3, is enriched at dSSRs. Blue bar arrows indicate the transcription direction in each polycistron. Genes in the same polycistron are stained with the same color. dSSRs, purple bars; cSSRs, yellow bars. Red bars represent H2B.V peaks (fold equal to 4 over the input, Poisson p-value over input required = 1.00e-04) obtained by HOMER in the epimastigote form. Rectangles with interrupted lines highlight dSSRs. Summary plot obtained by deepTools for H2B.V and H3 (ChIP/input) enrichment at polycistrons (B), dSSRs and cSSRs (C). Ten kilobases upstream and downstream of the abovementioned regions were also analyzed. All genomic features were automatically scaled as indicated (Start-End). Polycistron start and end is related, respectively, to the first ATG (from the first CDS) and the last stop codon (from the last CDS). dSSR and cSSR start refer to the first coordinate downstream, respectively, to the start and end of the left polycistron. dSSR and cSSR end refer to the first coordinate upstream, respectively, to the start and end of the right polycistron. D. Percentual distribution of H2B.V peaks (Fold 4) on all 41 chromosomes of the CL Brener Esmeraldo-like haplotype.

https://doi.org/10.1371/journal.ppat.1009694.g001

We used a peak calling algorithm to faithfully identify and compare peaks among life forms and genome features (S2B and S2D Fig and S2 and S3 Tables). Peaks were obtained for each replicate separately, but only the overlapping peak set was used for further analysis (S2B and S2D Fig). From the 265 peaks found in epimastigotes (fold change > = 4, Poisson p-value over input required = 1.00e-04), 91.3% were located at dSSR, and only 8.7% were found at non-SSR in the CL Brener assembly (Fig 1D and S2 Table). A higher enrichment of H2B.V at dSSR (76%) over non-SSR (24%) was also found in the TCC assembly (S3C Fig).

Although the great majority of H2B.V peaks were located within a dSSR, not all dSSRs contained H2B.V enrichment (S4A Fig). Specifically, 38% and 68% of dSSRs did not have any H2B.V peak (considering a fold change (ChIP/input) of 2) in the CL Brener and TCC assembly, respectively. However, these dSSRs usually flank monocistrons (67% and 60% in the CL Brener and TCC assembly, respectively); small (fewer than 4 CDSs) polycistrons (7.5% and 12% in CL Brener at TCC, respectively); snoDNA, tDNA (21% in the TCC assembly) (S5A Fig) or rDNA (S5B Fig). In the CL Brener assembly, we additionally noticed that 20% of the dSSRs that did not harbor the H2B.V peak were located close to a gap assembly region. In short, only 5% and 6% of dSSRs located between two protein coding polycistronic regions (with more than 4 CDSs) lacked H2B.V enrichment.

Among the H2B.V peaks located at non-SSRs, we found that they were closely associated with tDNA genes (25% and 5% in CL Brener and TCC, respectively), snoRNA genes (4% in CL Brener and 6% in TCC), and, more strikingly, between the disrupted and conserved genome compartments (46% -11 out of 23, in the CL Brener assembly) (Figs 1D, 2A, S3C and S4A). In the TCC assembly, 45% of the non-SSR H2B.V peaks were between disrupted and conserved compartments (S3C and S4B Figs).

To further explore this phenomenon, we evaluated the enrichment of H2B.V among multigenic family members composing the disrupted genome compartments (trans-sialidases, MASP, and mucins) and those that could be part of both conserved and disrupted compartments (GP63, RHS, and DGF-1). Heatmaps of H2B.V ChIP data indicated that members of disrupted families were more enriched in H2B.V (Fig 2B). Hierarchical clustering of all multigenic family members showed that mucins and MASP were preferentially enriched on H2B.V (14 and 15%, respectively, considering each gene class individually) (S4B and S4C Fig). Interestingly, evaluation of these two gene families in the most recent genome assembly with long reads [21] suggested that they are preferentially located at the first CDSs of a given PTU, which in turn would partially explain the enrichment of H2B.V in these gene families.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. H2B.V enrichment is also found at the border of conserved and disrupted genome compartments and around tDNA loci.

A. IGV snapshot of a 120-kb region from Chr 4S highlighting H2B.V enrichment around tDNA loci (black arrow) and between conserved and disrupted genome compartments (rectangle with interrupted lines). Wig files from the ChIP H2B.V/input and ChIP H3/input were obtained by COVERnant. Blue bar arrows indicate the transcription direction in each polycistron. Genes in the same polycistron are stained with the same color. Black and red bars represent respectively, tDNA loci and H2B.V -peaks (fold equal to 4 over the input, Poisson p-value over input required = 1.00e-04) obtained by HOMER in epimastigote form. The conserved and disrupted compartments are represented by green and orange bars, respectively. B. Heatmaps of multigenic family members showing ratios of ChIP H2B.V/input and ChIP H3/input in two epimastigote replicates. Note an enrichment (red) of H2B.V at MASP, mucin and TS genes. C. Summary plot obtained by deepTools of the normalized (ChIP/input) H2B.V reads at tDNA loci and the adjacent 2 kb upstream and downstream regions. The tDNA locus was automatically scaled as indicated (first (S) and last (E) nucleotides of each locus). Normalized data were classified into 3 hierarchical clusters (dark blue– 13 genes, light blue– 11 genes and yellow -31 genes) based on the H2B.V enrichment pattern.

https://doi.org/10.1371/journal.ppat.1009694.g002

A considerable number of non-SSR H2B.V peaks were located close to tDNA genes (Figs 2A and S5A). Thus, we evaluated the enrichment of H2B.V in tDNA loci by deeptools analysis (Fig 2C). H2B.V was enriched either at upstream or downstream regions of tDNA genes, with no clear association with tDNA anticodon type or adjacent polycistron transcription direction. tDNA genes were present alone or in clusters of 2–10 genes totaling 21 loci in the CL Brener Esmeraldo-like haplotype. Most of them were located within a protein-coding polycistron, suggesting that tDNA loci might interfere with RNA Pol II transcription elongation. Of the 10 loci of tDNA located at non-SSRs, seven were associated with a H2B.V peak (fold >2); from the six tDNA loci located at dSSR, all were associated with an H2B.V peak, while 2 out of 4 located at cSSR had an H2B.V peak.

Taken together, H2B.V was enriched at dSSRs flanking protein-coding polycistrons, at some tDNA loci, and at regions between the disrupted and conserved genome compartments. In contrast, dSSRs mainly associated with monocistrons and small polycistrons did not have H2B.V enrichment.

H2B.V Chip-seq enrichment differs greatly among epimastigote and TCT forms

Previously, H2B.V was found to be enriched in TCT chromatin when similar masses (in micrograms) of chromatin proteins released upon DNase treatment, from both life forms were compared [36]. We confirmed that H2B.V was enriched in TCT whole cell extracts (WCE) when equal numbers of epimastigotes and TCT parasites (in biological quadruplicates) were compared, while no important difference was found for histone H3 (S6A Fig).

Intriguingly, fewer sites of H2B.V enrichment were found along the TCT genome when compared with epimastigotes (Figs 1A, S2B and S3A). While 265 H2B.V-enriched peaks were found at epimastigotes, and only 6 peaks were found at TCTs (fold change over the input > = 4) in the CL Brener assembly. Using the same parameters, fewer and smaller (in width, in bp) peaks were found in TCT life forms than in epimastigote forms in the TCC assembly (S2B and S2C Fig). This finding agrees with lower H2B.V staining in myc-tagged TCTs nuclei detected by IFA (S1G Fig) but contrasts with previous findings [31].

Thus, we hypothesize that H2B.V may weakly interact with the TCT’s chromatin and be released upon experimental procedures like IFA and ChIP. To address this, we performed a chromatin sequential salt extraction assay, which evaluates the relative binding affinities of chromatin-associated proteins [37]. The results indicated that H2B.V (and H3) is released from chromatin in lower salt concentrations in TCTs than epimastigote forms, suggesting that H2B.V has a lower binding affinity to TCTs chromatin (S6B Fig).

As a whole, these findings suggest that the parasite modulates H2B.V enrichment in life forms, as will be discussed below.

H2B.V interacts with a bromodomain factor and nucleolar proteins

To gain better insights into the role of H2B.V in chromatin, we evaluated its interaction partners in epimastigotes and TCTs by pulldown assays. Therefore, recombinant H2B.V and canonical H2B (as a bait control) were used to obtain specific H2B.V interactors using the WCE of epimastigotes and TCTs. In addition, four control samples (described in detail in the Materials and Methods section) were also analyzed, which included the analysis of both parasite WCE and the recombinant protein extracts upon affinity resin incubation. From here, all experiments were performed with the T. cruzi Y strain for comparison with previous results [36].

Protein eluates were processed for mass spectrometry, and the results were subjected to in silico filtering steps (described in Materials and Methods) to obtain a more meaningful protein set of H2B.V interactors (Fig 3A and S4 and S5 Tables). From the final list of 39 interactors, two proteins were present in all pulldown eluates: double RNA binding domain protein 3 (TcCLB.506649.80) and polyadenylate-binding protein (TcCLB.508461.140), which is known to interact selectively and noncovalently with a sequence of adenylyl residues in an RNA molecule and participate in mRNA maturation at the poly(A) tail [38]. Interestingly, H2B.V and H2B interacts more with proteins found at TCTs than at epimastigote extracts. Eight proteins were found to interact with H2B.V in epimastigotes, whereas 26 proteins were found to interact with H2B.V in TCTs. Six were common to both, including 2 nucleosome assembly proteins (TcCLB.507963.79 and TcCLB.504839.50) and a hypothetical protein (TcCLB.509761.10) that was recently annotated as histone chaperone Rttp106-like, known in S. cerevisiae for being associated with DNA replication and heterochromatin silencing [39]. Interestingly, the latter was not found in canonical H2B pulldowns, suggesting that it might be associated with H2B. V nucleosome assembly.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. H2B.V in vitro interactors.

A. Venn diagram showing the number of common and specific interactors of H2B.V and H2B using epimastigote and TCT extracts. The total number of eluted proteins in each pulldown is shown between parentheses. Some relevant proteins are highlighted. B. BDF2 pulldown assays. Recombinant BDF2 tagged with HA and 6xHis was incubated with NTA-Ni resins with or without H2B.V-myc-tagged parasite extracts (epimastigotes or TCTs). Eluates were fractionated by 15% SDS-PAGE, and protein interactors were revealed by WB with anti-HA (against rBDF2), anti-myc (against H2B. V-myc tagged), and anti-H3 (as a chromatin control). C. H2A.Z interacts with H2B.V in vitro. H2A.Z-MBP and H2B.V-6xHis recombinants were incubated together or separately with amylose resin, known to specifically interact with MBP. Eluates were fractionated by 15% SDS-PAGE and stained with Coomassie. H2A.Z-MBP is approximately 60 kDa, and H2B.V-6xHis is 16.5 kDa. The * indicates a contaminant of lysed extraction.

https://doi.org/10.1371/journal.ppat.1009694.g003

Among H2B.V and H2B interactors in TCT extracts, we found a high mobility group protein TDP1 (TcCLB.507951.114), which recognizes and links noncanonical DNA structures, such as cruciform DNA and circular mini-DNA, in addition to its involvement in DNA folding [40,41]; a lupus La protein homolog (TcCLB.511367.60), which is an mRNA ligand and participates in mRNA maturation and translation, and stabilization of histone mRNAs during the S phase in humans [42,43], in addition to binding the 3′ poly(U)-rich in nascent RNA polymerase III transcripts participating in its correct folding and maturation in T. brucei [44]; a retrotransposon hot spot (RHS) protein (TcCLB.506113.60); and tousled-like kinase II (TcCLB.510597.9). The latter has enzymatic activity and is involved in several biological processes, such as DNA replication and S-phase progression, repair and H3S10 phosphorylation [45,46].

One of the most interesting findings was the presence of 3 nucleolar proteins specifically in H2B.V-TCT pulldowns: nucleolar protein 56 (TcCLB.511573.58), fibrillarin (TcCLB.510105.50), nucleolar RNA-binding protein (TcCLB.511573.58). In addition, two other hypothetical proteins (TcCLB.508409.180 and TcCLB.503829.80) were also detected, which were previously annotated as proteins of the nuclear pore and nuclear envelope, respectively [47,48]. Among H2B.V-specific interactors, we identified protein bromodomain factor 2 (BDF2) (TcCLB.506553.20), an epigenetic factor that recognizes histone acetylation [49,50]. In T. cruzi, BDF2 has already been shown to interact with histone H4 acetylated residues (K10 and K14), showing increased levels after exposure to UV in epimastigote forms and likely being involved in the response to DNA damage [51].

The H2B.V and BDF2 interaction was validated by reverse pulldown assays using recombinant BDF2-6xHis-HA and WCE of H2B.V-myc-tagged parasites (epimastigotes and TCTs) (Fig 3B). In contrast to what was observed in H2B.V pulldowns, here the BDF2 interacted with H2B.V in both life forms. In addition, the BDF2 interaction with H2B.V seemed to be specific and not a spurious histone interaction, as histone H3 was not eluted from the BDF2 pulldown. Nevertheless, BDF2 interaction with other core histones should be evaluated to confirm this specific interaction.

It is worth mentioning that besides H2B.V and H2B, we only detected H4 peptides in these pulldown assays. Here, the proteomics protocol was not optimized for histone identification, which is often overdigested by trypsin due to high amounts of lysine and arginine residues. In T. brucei, H2B.V dimerizes with the variant H2A.Z [52] and is preferentially detected at dSSRs. To verify whether T. cruzi H2B.V also dimerized with H2A.Z, we cloned and expressed the recombinant H2A.Z (TcCLB.511323.40) fused to a maltose binding protein (MBP) and incubated it with the 6xHis-tagged H2B.V recombinant (Fig 3C). Upon elution with maltose, both H2A.Z-MBP (~60 kDa) and H2B.V-6xHis (~16.5 kDa) were eluted, indicating that H2A.Z-H2B.V also interacted in T. cruzi. This in vitro interaction was further confirmed by Western blot analysis using polyclonal antibodies against H2A.Z (S7 Fig).

H2B.V is essential but H2B.V heterozygous knockout (H2B.V-HtzKO) epimastigotes do not display alterations in growth, replication or global transcription levels

To further understand the biological role of histone H2B.V in T. cruzi, we generated H2B.V knockout parasites by homologous recombination. Therefore, constructs containing neomycin resistance genes flanked by sequences from the 5’ and 3’ UTRs of the H2B.V gene (NEO cassette) were generated (S8A Fig) and used to obtain parasites that had lost one H2B.V allele and gained resistance to neomycin. The correct insertion of the neomycin cassette into the H2B.V locus was confirmed by PCR using specific primers (S8B Fig). Many attempts to delete the second H2B.V allele and generate double knockout (or homozygous knockout) parasites using a similar strategy described above (but with a cassette containing the hygromycin resistance gene–described in Materials and Methods) were performed without success. Thus, these results suggested that the H2B.V gene must be essential for T. cruzi, similar to T. brucei [52]. Thus, all phenotypic analyses described below were performed using two clones (cl 2 and 4) isolated from the transfected parasite population of HtzKO.

First, we confirmed that H2B.V transcription levels were decreased in HtzKO parasites by qPCR (Fig 4A) and by quantitative proteomics assays (iBAQ values) using basic WCE (S8C Fig). H2B.V-HtzKO cl2 and cl4 had, on average, 4 times less H2B.V protein than wild-type cells (unpaired t-test, p-value < 0.05). In accordance, the levels of histone H4 and all other histones did not change significantly in H2B.V-HtzKO mice. Then, we evaluated whether the lower levels of H2B.V in H2B.V-HtzKO parasites interfered with growth and cell cycle progression. No statistically significant differences were found among H2B.V-HtzKO clones and wild-type parasites when growth and cell cycle phases were evaluated (Figs 4B and S9A–S9C).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Phenotypic evaluation of H2B.V-HtzKO parasites.

A. H2B.V relative expression levels were evaluated by qPCR in wild-type and H2B.V-HtzKO (cl 2 and 4) parasites. H2B.V transcript levels were assessed relative to GAPDH levels (ΔCtH2BV–ΔCtGAPDH). One-way ANOVA: * p-value < 0.05. B. Mean growth curves (in log10) for wild-type and H2B.V-HtzKO (cl 2 and 4) parasites. C. Metacyclogenesis assay. The number of metacyclic trypomastigotes, epimastigotes and intermediate forms (herein, classified as other) were counted in 96-h supernatant upon nutritional stress as described in the Materials and Methods. Values were normalized to their respective numbers in wild-type cells. Experiments were performed in biological triplicates. Unpaired T-test: *p < 0.05 and ***p < 0.01 with Bonferroni correction. D. H2B.V-HtzKO parasites are more infective. LLCMK2 cells were infected with TCTs (1:40; cells:parasite). Slides were fixed, and nuclei were labeled with DAPI. The number of infected cells was counted relative to the total cells after 1 h of infection by fluorescence microscopy. Unpaired t-test with Welch’s correction *** p < 0.0001.

https://doi.org/10.1371/journal.ppat.1009694.g004

As H2B.V was enriched at putative TSRs, we asked whether H2B. V-HtzKO parasites would show a decrease in global transcription by evaluating EU incorporation. Again, no significant global differences were found among clones and wild-type parasites (S9D Fig).

The reduced H2B.V levels might result in a more stable chromatin structure once nucleosomes containing H2A.Z-H2B.V are more unstable than canonical nucleosomes [13]; thus, we evaluated whether H2B.V-HtzKO parasites would have global changes in chromatin structure by evaluating the amount of active chromatin extracted by the FAIRE approach [53]. H2B.V-HtzKO parasites and wild-type parasites have similar levels of open chromatin on H2B.V-HtzKOs (S9E Fig).

Since we evaluated global effects on transcription and active chromatin levels, we could not rule out that specific genomic regions could be affected in H2B.V-HtzKO parasites. Future transcriptomic and sequencing analyses from the data shown in S9D and S9E Fig could clarify this issue.

Changes in H2B.V levels can affect metacyclogenesis and infection of mammalian cells

To assess whether the decrease in H2B.V levels in H2B.V-HtzKO parasites affect the differentiation of epimastigotes into metacyclic forms, we performed a metacyclogenesis assay evaluating both the presence of fully differentiated metacyclics and their intermediate forms (based on kinetoplast morphology and position relative to the nucleus as previously described [25]. Labeling of the metacyclic membrane markers GP90 and GP82- was also used to discriminate differentiated parasites [54].

In the stationary growth phase, H2B.V-HtzKOs (cl2) had approximately 60% more intermediate parasite forms than wild-type parasites (S9F Fig), suggesting that changes in H2B.V levels might facilitate the differentiation of epimastigotes into metacyclic trypomastigotes. Furthermore, in the 96-h supernatant after the induction of metacyclogenesis, almost 3 times more trypomastigote metacyclics were found in H2B.V-HtzKO when compared to wild-type parasites (Fig 4C). This supernatant was also enriched in non-metacyclic parasites (classified as “others”, which included epimastigote and intermediate forms), suggesting that changes in H2B.V levels promoted loss of adhesion to substrates.

We also evaluated whether the ability to infect mammalian cells would be affected in H2B.V-HtzKOs TCTs. Surprisingly, we found that cl2 and cl4 infected approximately twice as many cells as wild type (Fig 4D). Taken together, these data suggested that decreasing H2B.V levels might interfere with parasite differentiation and infection of mammalian cells.

Parasites, metacyclogenesis, growth curves and mammalian cell invasion

T. cruzi epimastigotes (Y and CL Brener strains) were cultured at 28°C in LIT medium supplemented with 10% fetal bovine serum (FBS—Vitrocell), glucose 0.4%, hemin 0.1 μM and penicillin-G 59 mg/L as previously described (Camargo, 1964). H2B.V-HtzKO (cl 2 and 4- Y strain), H2B.V-myc tagged (CL Brener strain) epimastigote forms were maintained in the abovementioned medium supplemented with 500 μg/mL G418 (Gibco) and 500 μg/mL puromycin. For growth curves, epimastigotes were diluted to 5 × 10⁶ cells/mL and monitored for four days. Cell density was counted in the Z2 Coulter Particle Count and Size Analyzer (Beckman Coulter) using a 100-μm filter. TCT forms (Y and CL Brener strains) were obtained from the supernatant of infected LLCMK2 cells after 1 week of infection (1:40 –cell:parasites) according to [78]. Cells were maintained in DMEM (Gibco) containing NaHCO₃ 3.7 g/L, penicillin G 0.059 g/L, streptomycin 0.133 g/L and FBS 10% at 37°C with 5% CO2. For mammalian cell invasion assays, after 1 h of infection (1:40, cell:parasites), cells were fixed with 4% paraformaldehyde, stained with DAPI and visualized by Olympus IX81 fluorescence microscopy. For ChIP-seq experiments, TCTs samples with a maximal of 5% of amastigotes were used. For Y strain (Wt and H2B.V-HtzKOs) the metacyclogenesis protocol described by [79] was followed by evaluating the TAU 3AAG culture supernatant upon 96–144 h of nutritional stress. For CL Brener (H2B.V-myc tagged), a culture of ~5 x10⁶ epimastigote/mL was maintained undisturbed for 8 days in RPMI medium without FBS, following the protocol described by [80]. Intermediate forms were classified based on kinetoplast and nucleus position, as shown in [25]. Antibodies against GP90 and GP82 were used to facilitate visualization of metacyclic parasites. Parasites were counted in technical triplicates from each biological triplicate.

Generation of H2B.V-HtzKO parasites

All primers sequences are described in S1 Table. The regions upstream and downstream of the H2B.V gene (TcCLB.506779.150) were cloned from the genomic DNA of the T. cruzi Y strain into the recombinant plasmids pTc2KO-neo and pTc2KO-hyg [81]. The upstream region of the H2B.V gene (434 bp) was amplified by PCR using the primers H2B.V_KpnI (forward) and H2B.V_SalI (reverse) (fragment 5’flank_H2B.V). Simultaneously, the 507-bp fragment from the intergenic region downstream of the H2B.V gene was amplified by the primers H2B.V_BamHI (forward) and H2B.V_XbaI (reverse) (fragment 3’flank_H2B.V). The recombinant plasmids pTc2KO-H2B.V-neo and pTc2KO-H2B.V-hyg were purified, and the “NEO cassette” (2,671 bp) and “HYG cassette” (2.971 bp) were amplified by PCR using the primers H2B.V_KpnI (forward) and H2B.V_XbaI (reverse) (S8A and S8B Fig left). Parasite transfection and selection were performed by homologous recombination of the NEO cassette, and parasites resistant to G418 were confirmed by PCR using the primers NEO_F and NEO_R; H2B_EXT_FOR and H2B_EXT_REV, NEO Out R and NEO Out F (S8B Fig right). The neomycin-resistant parasites were electroporated with the “HYG cassette” (5’FLANK-HYG-3’FLANK) to generate the H2B.V homozygous knockout, which would be resistant to both G418 and hygromycin B (at 500 μg/mL). Clones were obtained by serial dilution by FACS sorting.

Generation of H2B.V-myc parasites

All primer sequences are described in S1 Table. The CRISPR-Cas9 protocol was based on [82]. Donor DNA and sgRNA were amplified by PCR using high-fidelity Phusion DNA Polymerases (Thermo Fisher Scientific) considering the H2B.V (TcCLB.506779.150) locus. Primers pMOtag forward and reverse were used to amplify the donor DNA from the pMOtag-23M plasmid [83], and the final PCR product was composed of 30 bp of the homologous arm (final 30 nucleotides from the H2B.V gene, excluding the stop codon), 3 copies of Myc epitope sequence, and the puromycin resistance gene and 30-bp homology arm (30 nucleotides from the intergenic region adjacent to the cleavage region). The donor DNA was inserted in the 3’ region of the H2B.V gene to be expressed at the C-terminal portion of H2B.V protein. Primers “TcH2Bv 3´ UTR sgRNA” and “sg scaffold” were used to amplify sgRNA. PCR products were purified (QIAquick PCR Purification Kit Protocol) and transfected into epimastigotes expressing Cas9 (CL Brener strain). A total of 10⁸ epimastigotes were resuspended in 350 μL of transfection buffer (90 mM sodium phosphate, 5 mM potassium chloride, 0.15 mM calcium chloride, 50 mM HEPES, pH 7.2) together with 50 μL of purified DNA (PCR products for donor DNA and sgRNA). The 400-μL final volume was placed in 0.2-cm cuvettes (Bio-Rad) and electroporated in the Bio-Rad Gene pulser with 2 pulses of 500 μF and 450 V. After electroporation, parasites were placed in fresh LIT medium with 10% SFB at 28°C. Then, after 24 h, puromycin was added to the medium at a final concentration of 10 μg/mL. After approximately two weeks, the polyclonal population was selected and then cloned. Epimastigotes were diluted to 1 parasite per mL in LIT/10% SFB containing puromycin and then plated in 96-well plates (200 μL per well). The parasites were incubated at 28°C with 5% CO_2, and monoclonal populations were obtained within approximately 3 weeks. Clones were obtained by serial dilution. Genome edition and allele-specific insertion were confirmed by PCR using the primers Fw_mycH2BVcheck_Chr27P (1), Fw_mycH2BVcheck_Chr27S (2), Rv_mycH2BVcheck_myc (3), and Rv_mycH2BVcheck_puromicin (4) (S1 Table).

Cloning and recombinant protein expression

All primer sequences are described in S1 Table. The H2B gene (TcCLB.511635.10), BDF2 (TcCLB.506553.20) and H2A.Z (TcCLB.511323.40) from T. cruzi (CL Brener strain) were amplified using the following pair of primers (all containing Eco RI and Hind III cleavage sites): H2Bc_F and H2Bc_R; HA_BDF2_Forward and BDF2_Reverse, and H2A.Z_Forward and H2A.Z_Reverse. BDF2 was amplified by fusion with HA from the vector pDEST17, which was kindly provided by Dr. Esteban Serra. The PCR products were purified and cloned first at pJET1/2blunt (Invitrogen), and after SANGER sequencing confirmation, the insert was subcloned either at pET28(a)+ (Novagen) (for H2B and BDF2) or at pMAL-p2x (NEB) (for H2A.Z). Plasmids were transfected into E. coli DH5α and/or at BL21 DE3 for recombinant expression as described previously [36]. Briefly, E. coli BL21 DE3 transformed either with pET28a (+)-H2B.V [36], pET28a(+)-H2B, pET28a(+)-BDF2 or pMAL-p2x-H2A.Z were inoculated in LB base medium (tryptone 1%, NaCl 1%, yeast extract 0.5%, pH 7) containing kanamycin 50 μg/mL at 37°C for up to 18 h, and recombinant expression was induced with 1 mM IPTG. H2B, H2B.V and BDF2 (His-tag) and H2A.Z (fused with a maltose-binding protein -MBP) purification were performed by affinity chromatography using Ni-NTA agarose (Qiagen) and amylose resin (NEB), respectively, according to the manufacturer’s recommendations.

Cell cycle and EU (5-Ethynyl-2’-uridine) assay

Epimastigotes in the exponential growth phase were fixed in 70% ethanol overnight at -20°C and treated with 10 μg/mL RNase A (Thermo Scientific) and 40 μg/mL propidium iodide (Sigma-Aldrich, St. Louis, MO). Cells were analyzed using the Attune® Acoustic Focusing Cytometer (Applied Biosystems). Epimastigotes in the exponential growth phase (~5x10⁶ cells/mL) were incubated with EU 10 μg/mL (Life) and fixed in cold 50% ethanol for 20 min, as described in [84]. Cells were analyzed using the Attune® Acoustic Focusing Cytometer (Applied Biosystems). Data from 20,000 events were analyzed using FlowJo v. 10.7.1 software.

Protein extracts

A total of 5 × 10⁸ parasites were used to obtain basic protein extracts [33]. Extracts were quantified using the Pierce BCA Protein Assay Kit protocol (Thermo Scientific) and stored at -20°C.

Pulldown assays

The pulldown protocol was based on the principle of immobilization of recombinant protein using Ni-NTA agarose resin (Qiagen). The experiments were performed in biological triplicates and designed to contain 4 control samples (resin Ni-NTA with i. bacteria extract, ii. Recombinant H2B. V or BDF2, iii. epimastigote extracts, iv. TCT extracts) and 2 experimental extracts (Ni-NTA resin i. epimastigote extracts and recombinant H2B.V or BDF2; ii. TCT extracts and recombinant H2B.V or BDF2). Recombinant protein was first incubated with nickel resins for 30 min under slow stirring at 4°C, followed by a 1-h incubation with parasite basic protein extracts (described above). The resin was washed three times with TBS, and recombinant protein and its interactors were eluted with 50 mM sodium phosphate, 0.3 M NaCl and 290 mM imidazole buffer. One-third of the eluate was reserved for fractionation by SDS-PAGE, and the remainder was reserved for proteomics analysis. The BDF2 eluates were analyzed by western blotting.

Proteomics analysis

Pulldown eluates and parasite extracts (basic protein extracts from H2B.V-HtzKO parasites) were precipitated with 20% TCA and resuspended in 8 M urea, 75 mM NaCl, and 50 mM Tris, pH 8.2. Protein was reduced with 5 mM DTT and alkylated with 14 mM iodoacetamide followed by digestion with trypsin (1:100, w:w) (Promega) as described in [85]. The peptides were resuspended in 0.1% formic acid and injected (5 μl) into a precolumn (Thermo—C18 of 5 μm, 2 cm x 100 μm) coupled to a nano HPLC (NanoLC-1DPlus, Proxeon). The separation was carried out in a capillary precolumn (10 cm x 75 μm containing C18 resins of 3 μm) in a 2–50% gradient of acetonitrile in 0.1% formic acid for 1 h. The eluted peptides were analyzed directly on an Orbitrap spectrometer (LTQ-Orbitrap Velos-Thermo). The 10 most intense ions were selected for fragmentation by CID. All ions were analyzed in positive ionization mode. The data in * raw format were processed using the Andromeda-MaxQuant program [86] with TriTrypDB (T. cruzi taxonomy—obtained from http://tritrypdb.org/tritrypdb/). Database searches were performed using the following parameters: carbamidomethylation of cysteine as a fixed modification; oxidation of methionine, N-terminal acetylation as variable modifications; tolerance MS1 of 6 ppm and MS2 of 0.5 Da, and 1% FDR. The ProteinGroups.txt output was sequentially processed using the Perseus program [86]. Proteins with LFQ intensity values equal to zero were considered absent. Proteins considered by MaxQuant as contaminants, presented in the reverse database and identified only by a modified peptide, were removed. For pulldown analysis, we also removed proteins presented (LFQ value > 0) in only one of the 3 biological replicates; proteins with a ratio of LFQ values from the experimental sample (EpiH2B, TripoH2B, EpiH2BV and TripoH2BV) to the control sample (Control_Epi, Control_TCT) lower than 1.5 (to ensure that eluted protein was enriched in the pulldown assay over the control samples); and common contaminant proteins (ribosomal, cytoskeleton, heat-shock and metabolic proteins). Finally, we only considered proteins that were present in a previous chromatin study [36] and those with T. brucei protein orthologs located in the nucleus/chromatin according to TrypTag. For histone quantitation on H2B.V-HtzKOs, the iBAQ values for all detected histones were considered 100%.

ChIP-seq

ChIP-seq experiments were performed as described in [16] with a few modifications. Epimastigote and TCT H2B.V-myc-tagged cells (1x10⁸) were cross-linked with 1% formaldehyde for 20 min at RT and quenched with 2 M glycine followed by washes with cold TDB buffer (5 mM KCl, 80 mM NaCl, 1 mM MgSO₄, 20 mM Na₂HPO₄, 2 mM NaH₂PO₄, and 20 mM glucose (pH 7.4)). The permeabilization buffer and NP-S buffer described in [16] were substituted for lysis buffer (1 mM potassium L-glutamate, 250 mM sucrose, 2.5 mM CaCl₂, 1 mM PMSF) [36]. Upon crosslinking, cells were resuspended in 1 mL lysis buffer, centrifuged, and resuspended in lysis buffer containing 0.1% Triton X-100 for 15 min at RT. Then, the cells were pelleted, washed, and incubated with 75 U of Mnase (Sigma Aldrich, #N3755) at 25°C for 10 min. After the addition of 10 mM EDTA to block digestion, the supernatant was collected and washed with lysis buffer with 0.1% SDS. The pellet was sonicated in a Covaris S220 (1 min, 10% duty factor, 200 cycles per burst). Sheared DNA was centrifuged (10,000 x g for 10 min at 4°C). Fifty microliters of Dynabeads protein G (Thermo Fisher) was resuspended in PBS-Tween (0.02%) containing 10 μg of purified anti-Myc antibody (Helmholtz Zentrum München, Monoclonal Antibody Core Facility) and incubated with slow rotation at 4°C overnight. Antibody-coupled beads were incubated at 4°C overnight with slow rotation with the MNase supernatant obtained above, washed eight times with cold RIPA buffer (50 mM HEPES-KOH pH 7.5, 500 mM LiCl, 1 mM EDTA, 1% NP-40, 0.7% Na-deoxycholate), washed one time with TE buffer containing 50 mM NaCl and eluted with 50 mM Tris-HCl pH 8.0, 10 mM EDTA, 1% SDS at 65°C for 30 min. The crosslinking of ChIP and input samples was reversed with 300 mM NaCl (at 65°C for 9 h) and treated with RNase A and proteinase K as described. After DNA purification, ChIP and input DNA samples were used to construct an Illumina library using TrueSeq adapters according to [16]. The libraries were sent for Illumina NextSeq sequencing with 75-bp paired-end sequencing at the Core Unit Systems Medicine, University of Würzburg.

ChIP-seq analysis

Quality filtering and adapter trimming of the Illumina sequencing reads were performed with fastp version 0.20.0 [60]. Reads with more than 5 N bases were discarded, as were those with more than 40% of their bases with Phred quality values below 15 (< Q15). All nucleotides from the 5’ or 3’ ends of each read with quality inferior to 5 (< Q5) were removed. Since these data were paired-end, we activated the correction algorithm: for size overlaps of at least 30 nucleotides (accepting 5 mismatches), a base is corrected if it is ultralow quality (< Q15), while the corresponding base at the mate read is high quality (> Q30). After the application of all these filters, reads shorter than 15 base pairs were removed. Approximately 94% of the reads passed the described filters and were used for downstream analysis. Quality control was performed with FastQC version 0.11.8 [61]. The local mapping of the filtered reads against the genome of the CL Brener Esmeraldo-like strain (release 32) and against the TCC strain (release 44) from TriTrypDB was performed with bowtie2 version 2.3.5.1 [62], with parameters more restrictive than the preset option: very-sensitive-local parameters (-D 25 -R 4 -N 1 -L 19 -i S,1,0.40—nceil L,0,0.15). Coverage plots were generated using COVERnant v.0.3.2 (https://github.com/konrad/COVERnant) and visualized using IGV (http://software.broadinstitute.org/software/igv/). COVERnant ratios were obtained by pairwise analysis of each ChIP and input sample. For the identification of genomic regions enriched in H2B.V, the bam files were analyzed using the findPeaks (-sytle histone), getDiffPeaks (-style histone) and mergePeaks (-d given [default] = maximum distance to merge) commands from HOMER [64] with default options. Peaks were obtained by the findPeaks command considering a fold (-f) enrichment of 2 or 4 over the corresponding input sample. To analyze histone enrichment at specific genomic regions, genomic coordinates of each feature were obtained from an in-house customized GFF file. Summary plots and heatmaps were obtained using the computeMatrix (with scale-regions and skipZeros options) and plotHeatmap (with hierarchical clustering) functions from deepTools2 [87,88]. The GFF file from T. cruzi (available at https://tritrypdb.org/) was customized in-house to contemplate the polycistrons, dSSRs and cSSRs.

WB and immunofluorescence (IFA)

WB and IFA were performed as described with antibodies, and the dilutions used were as follows: anti-Myc (Cell Signaling) 1:3000; anti-HA (Cell Signaling) 1:3000, anti-H3 (Abcam) 1:10000, anti-BIP 1:1000 (polyclonal), anti-H2A. Z (polyclonal–kindly provided by Dr. Nicolai Siegel), anti-His Tag 1:3000 (Cell Signaling); and anti-GAPDH 1:4000 (polyclonal) [89]. For IFA of myc-tagged parasites, after fixation with 4% paraformaldehyde and permeabilization with Triton X-100, an overnight incubation with anti-Myc (Cell Signaling) antibody 1:3000 in 1% PBS/BSA at 4°C followed by secondary antibody conjugated to Alexa Fluor 555 (Thermo Fisher) at 1:500 in PBS/BSA 1% was performed. When mentioned, IFA using anti-GP90 and anti-GP82 1:1000 (kindly provided by Nobuko Yoshida) was performed as described previously [90]. Slides were further mounted with Vectashield plus DAPI (Vector laboratories) and visualized at 100 X magnification in an inverted OLYMPUS microscope model IX81 with Z axis motorization. Images were acquired using OLYMPUS CELL R version software 3.2.

Quantitative PCR

TRIzol Reagent (Invitrogen) was used to isolate total RNA from T. cruzi epimastigotes (Y strain–wt and HtzKO cl2 and cl4; and CL Brener strain–wt, Cas9, H2B.V-myc tagged) in exponential growth phase. Next, 10 μg from each total RNA sample was subjected to DNAse I (Sigma) treatment to remove genomic DNA contamination. Then, 1 μg of purified RNA was used for cDNA synthesis with a Super Script Reverse Transcriptase Kit (Invitrogen) following the manufacturer’s recommendations. Quantitative real-time PCRs were performed using H2B.V_qPCR_F and H2B.V_qPCR_R as the target genes, and GAPDH_Tcruzi_qPCR_Fw and GAPDH_Tcruzi_qPCR_Rv were used as endogenous control genes (S1 Table). For each reaction, 5 μL of 2X Power SYBR Green PCR Master Mix (Applied Biosystems), 2 ng cDNA from wt and HtzKOs (Y strain); and 10 ng cDNA from wt, Cas9 and H2B.V-myc tagged (CL Brener strain); and 300 nM H2B.V primer or 100 nM GAPDH primer were used. All reactions were carried out in triplicate in a 96-well plate. The plates were run on a StepOnePlus Real Time PCR System (Applied Biosystems). The relative quantification of H2B.V transcripts was calculated by the 2^-ΔCt method [91].

Faire assay

The FAIRE samples were prepared as described in [92] with some modifications. The 2x10⁸ epimastigotes of wild type, H2B.V HtzKO cl2 and cl4 were fixed with 37% formaldehyde and directly added to LIT medium to a final concentration of 1% for 5 minutes at RT. The formaldehyde was quenched with 2.5 M glycine to a final concentration of 125 mM. Then, the parasite pellets were resuspended in 2 mL of cold TELT buffer (50 mM Tris-HCl pH 8.0, 62.5 mM EDTA, 2.5 M LiCl, 4% Triton X-100) and sonicated in a Tommy Ultrasonic Disruptor UD-201 apparatus with an output of 4 and duty of 30 for 10 cycles of 30 seconds with 1-minute intervals. To prepare input control DNA, one-fourth (500 μL) of each sample was incubated with 5 μL of proteinase K for 1 h at 55°C, and the cross-linking was reversed by overnight incubation at 65°C. After that, the DNA was extracted with phenol:chloroform:isoamyl alcohol solution and precipitated with 3 M sodium acetate (pH 5.2) and 95% ethanol. The input control DNA obtained was quantified using a NanoDrop (Thermo Fisher). To verify the sonication efficiency, 500 ng of input DNA control was run on a 1% agarose gel. To prepare FAIRE DNA, half of the fixed cell lysate (1 mL) was used. DNA was extracted as described above; however, protein removal by proteinase K and cross-linking reversal were performed after DNA purification. Then, DNA was purified with a MinElute PCR Purification Kit (Qiagen) following the manufacturer’s recommendations. To assess the DNA yield, the total quantity of control DNA or FAIRE DNA (in ng, obtained by Nanodrop quantitation) was divided by the volume of cell lysate used for control DNA or FAIRE DNA obtention (in μL), respectively. Considering the control DNA yield as 100%, the retrieval ratio of FAIRE DNA for each sample was calculated.

Sequential salt extraction of chromatin proteins

10⁸ epimastigotes and TCTs H2B.V-myc tagged (CL Brener strain) were collected and lysed with extraction buffer (0.1% Triton X-100, 10 mM Tris-HCl [pH 7.4], 100 mM NaCl, 300 mM sucrose, 3 mM MgCl₂, 50 mM NaF, 1 mM Na₃VO₄, 0.5 mM phenylmethylsulfonyl fluoride, and EDTA-free Complete protease inhibitor cocktail [Roche]), described at [93] for 10 min at 4°C under agitation. The samples were pelleted, and the supernatants were saved as soluble fraction (soluble fraction 1 –SF1). This step was repeated (soluble fraction 2 –SF2). The pellet was then treated with a sequential salt extraction gradient of 0, 200, 400, 600 mM de NaCl in mRIPA buffer, and the supernatants collected as described at [37]. The extraction was performed in biological duplicates and submitted to Western blot. The protein extracts equivalent to 200 ng of DNA (measured by Qubit) for Epimastigotes and TCTs were fractionated by SDS-Page 12%.