Nucleosome modifications, together with specific proteins recruited to these modifications (histone readers), dictate many fundamental chromatin-associated processes in eukaryotes. This field is now emerging as a fascinating research area in Plasmodium
, and is clearly linked to virulence gene control in this organism. Here, we have performed an in depth analysis of histone phosphorylation of asexual blood stage parasites of P. falciparum
. To this end, we have developed improved methods of extracting histone samples that retain unprecedented levels of PTMs. Our analysis of phospho-enriched histone peptides revealed multiple phosphorylation sites mostly at the N-terminal region of most histones. These marks are frequently seen in combination with neighbouring lysine acetylation (and methylation). In addition, we identified Pf14-3-3I as a phospho histone mark binding protein. Previously, we and others had identified heterochromatin protein 1 (PfHP1) binding to H3K9 methylated as a key mediator in heterochromatin formation linked to the expression of clonally variant gene families 
. Pf14-3-3I is the second P. falciparum
histone mark reader protein to be identified.
Phosphorylation of histones plays a role in cell signalling and transcriptional regulation in a number of eukaryotic organisms (reviewed in 
). Plasmodial histones contain abundant serine, threonine and tyrosine residues for potential phosphorylation. Although previous studies have identified the role of histone methylation and acetylation in plasmodial gene regulation, histone phosphorylation was not reported in these studies 
. In these studies, traditional methods of acid extraction were used to obtain partially purified proteins for further phospho-protein analysis. 
. However, the labile nature of phospho-marks and the relatively low abundance of most phospho-modifications may explain the negative results in previous reports on histone marks 
. For this reason, we combined improved purification methods of histones with phosphopeptide enrichment to revisit this topic 
. We improved on two traditional histone extraction protocols, namely acid extraction and non-acid high-salt extraction 
, to better preserve PTMs including phosphorylation. Using commercially available antibodies we were able to demonstrate the retention of various phospho-modifications in the histone samples prepared by either method. All samples were initially analyzed by LC-MS/MS, without enriching for phosphopeptides. This step enabled us to identify many PTMs with a significant mascot score, which were not manually validated (data not shown). We were also able to identify multiple modifications on the same peptide, which supports a possible crosstalk between distinct histone marks in vivo
. At this level, we were able to identify only three, probably the most abundant phospho-modified residues for both H3.1 and H3.3, namely Ser-28, Ser-32, and Thr-45 (data not shown). Subsequent experiments included phosphopeptide enrichment prior LC-MS/MS analysis. This led to a dramatic increase in the number of detected phosphorylation sites specific to P. falciparum
histones ( and S1
). Two very recent studies analysed the general phosphoproteome of P. falciparum
and one of these studies reported several histone phosphorylation marks in late schizonts 
. Only a fraction of these reported modifications overlap with the phospho marks identified in the present work ( and S1
). Conversely, other modifications reported only by that study were also identified in our LC-MS/MS analysis but did not pass our rigorous filter (see Experimental Procedures). It remains unclear if the differences observed in both studies are due to the fact that late schizont parasites show a distinct histone phospho-marks compared to younger parasite stages (rings and trophozoites in this study) or is due to different protein extraction methods.
Histone modifications can be recognized by nonhistone proteins with domains specific for methylated lysines, acetylated lysines or phosphorylated serines. These histone readers can recruit other proteins such as histone-modifying enzymes and promote different biological outcomes for the DNA linked to this chromatin region. Different histone phosphorylation binding modules such as 14-3-3 and BRCT are known. The 14-3-3 proteins have been reported to bind histone H3 phosphorylated at Ser-10 and/or Ser-28 residues. They have been implicated in diverse roles of regulating chromatin remodeling, transcription activation and hence gene regulation, though the exact mechanism behind these function remains elusive (reviewed in 
). Based on our finding of frequent histone phosphorylation in P. falciparum
, we searched for plasmodial proteins containing 14-3-3 domains, which have been reported to bind histone H3 phosphorylated at Ser-10 and/or Ser-28 residues. Two putative 14-3-3 proteins are highly expressed during the asexual parasite cycle 
. Sequence alignment of these proteins to 14-3-3 proteins from other model organisms indicated conservation of residues important for the interaction of these proteins with phosphorylated proteins or peptides. Residues responsible for dimer formation are also completely conserved for Pf14-3-3I, but only partially conserved for Pf14-3-3II () 
We also detected H3S10ph and H3S28ph modifications in our purified samples. Hence, we produced GST-tagged recombinant versions of these two putative parasite 14-3-3 proteins, which we named Pf14-3-3I and Pf14-3-3II, to analyse their histone binding properties. In an ELISA-based binding assay, both of these proteins bound purified parasite histones, with 14-3-3I binding distinctively to the H3S28ph peptide, and not the H3S10ph peptide, demonstrating selectivity for one phospho-mark over the other. The canonical 14-3-3 binding motif includes a proline residue at position P+2 from the phosphoserine 
. This proline introduces a turn in the phosphopeptide that allows the remaining protein to exit the 14-3-3 binding pocket 
. Both P. falciparum
histone H3.1 and H3.3 have a proline at P+2 after S28, the canonical 14-3-3 binding motif. Conversely, P. falciparum
histone H3.1 and H3.3 contain motifs ARKSTAG and ARKSTGG, respectively, in the vicinity of S10. The GG sequence in H3.3 could allow such a turn but the AG sequence in H3.1 might not allow a similar degree of flexibility. The differences in these potential 14-3-3 binding motifs might explain the differential Pf14-3-3 histone/peptide binding results. In yeast, 14-3-3 proteins preferentially recognize H3S10phK14ac over H3S10ph 
, demonstrating how neighbouring modifications can affect binding of these proteins. Hence, we tested neighbourhood effect by including dually modified peptides H3S10phK14ac and H3S28phS32ph in the study. Recombinant 14-3-3I clearly bound H3S28phS32ph peptide but did not bind to H3S10phK14ac peptide in our in vitro
studies. This demonstrated that neighbouring modifications did not affect the binding pattern of the protein. However, even though Pf14-3-3II bound purified parasite histones, it did not preferentially bind to any of the H3 peptides included in this study, while demonstrating low level of binding to all the peptides. This leads us to speculate that Pf14-3-3II might be recognizing some specific combination of modifications retained in purified parasite histones that are not represented in our synthetic peptides. Alternatively, as the residues responsible for dimer formation are only partially conserved for Pf14-3-3II it remains possible that this protein functions as a heterodimer with either Pf14-3-3I or other presently unidentified proteins. The 14-3-3 proteins are known to function as both homo- and heterodimers 
. Further experiments need to be done to confirm whether Pf14-3-3II is another member of the histone mark reading machinery and to what extent, if any, protein dimerization plays a role in that function. Additionally, these Pf14-3-3 proteins may be subject to a structure-based auto-inhibitory mechanism. Structural modelling using the I-TASSER server resulted in predicted Pf14-3-3 structures that contain C-terminal protein segments located in the canonical 14-3-3 phosphopeptide binding site (Figure S2
), as has been shown for the recently solved C. parvum
14-3-3 protein 
. This C-terminal region has been implicated in interfering with 14-3-3 ligand binding through folding back into the peptide binding pocket, providing a regulatory mechanism of 14-3-3 effector function 
. Strikingly, all five predicted structural models of Pf14-3-3II included a portion of their C-terminus in the phosphoprotein binding-pocket.
14-3-3 proteins are involved in the regulation of subcellular localization, activation or inhibition of enzymes, and signal transduction 
. Consistent with this pleiotropic role, immunolocalization analysis located Pf114-3-3I in cytoplasmic and nuclear compartments. Additionally, rodent malaria 14-3-3 proteins have been shown to interact, in a phospho-dependent manner, with the internalized host skeletal protein dematin and it might determine the localization of host-derived dematin inside the parasite 
. To further explore the biological role of Pf14-3-3 proteins, co-immunoprecipitation experiments may identify their interaction partners and chromatin immunoprecipitation assays may determine the chromatin occupation sites of these proteins and reveal a functional link to gene transcription or cell division.
In conclusion, our data set the framework for studies on histone phosphorylation mediated regulatory processes in chromatin biology of malaria parasites. This work opens up avenues to study signal transduction cascades leading to histone phosphorylation and ultimately controlling transcription and other nuclear processes in malaria parasites.