Phosphorylation of histone H4 on one or both of its two histidine residues has been known to occur in liver cells for nearly 40 years and has been associated with proliferation of hepatocytes during regeneration of the liver following mechanical damage. More recently, large increases in histone H4 histidine kinase activity have been found to occur associated with proliferation and differentiation of liver progenitor cells following chemical damage that prevents hepatocyte proliferation. In addition, it has been shown this histone H4 histidine kinase activity is elevated nearly 100-fold in human foetal liver and several hundredfold in hepatocellular carcinoma tissue compared with normal adult liver. In the present paper, we review what is currently known about histone H4 histidine phosphorylation, the kinase(s) responsible and the phosphatases capable of catalysing its dephosphorylation, and briefly summarize the techniques used to detect and measure the histidine phosphorylation of histone H4 and the corresponding kinase activity.
The key role of protein phosphorylation in the regulation of many aspects of cellular function has long been recognized. The characterization of serine/threonine kinases and their substrates preceded the recognition that phosphorylation of proteins also occurs on tyrosine residues via the action of the corresponding tyrosine kinases. Integral to these regulatory systems are the corresponding phosphatases that catalyse the hydrolysis of the phosphoester bonds in phosphoproteins, resulting in their dephosphorylation. In the early 1990s, the role of histidine phosphorylation and the two-component histidine kinases in bacterial sensor signalling was recognized. Later, analogous systems were also found in plants and fungi, but there is no evidence that such systems occur in animals. Histidine phosphorylation of animal proteins has, however, been long known to occur, and Boyer and co-workers were among the first to report it in mitochondrial proteins [1,2].
The detection and analysis of histidine phosphorylation is made more difficult than that of serine, threonine and tyrosine phosphorylation because of the acid-labile nature of the P-N bond in phosphohistidine and the fact that there are two phosphorylation sites on the imidazole ring of histidine, N-1 and N-3 (Figure 1). Much effort has thus gone into developing methods of detection and analysis that are compatible with this labile nature of phosphohistidine and to distinguish between the two isoforms of phosphohistidine.
Structures of the isoforms of phosphohistidine (1- and 3-phosphohistidine) and two non-hydrolysable analogues of 3-phosphohistidine
In the 1970s, Smith and co-workers first reported the occurrence of an HHK (histone H4 histidine kinase) in the nuclei of rat liver cells [3–5] and Walker-256 carcinosarcoma cells . Since then, similar kinase activity has been reported in Physarum polycephalum , Saccharomyces cerevisiae , porcine thymus , human fetal liver cells , human hepatocelluar carcinoma cells  and rat pancreatic β-cells [11,12].
In the present paper, we summarize what is currently known about histone H4 histidine phosphorylation, HHKs and corresponding phosphatases and the methods currently available for the detection and analysis of histone H4 phosphorylation.
Histone H4 histidine kinases
As described above, HHK activity has been found in a wide range of source tissues and cells. HHK activity seems to be associated with proliferating cells; indeed, in adult rat and especially adult human liver cells, HHK activity is very low [4,10]. However, following partial hepatectomy of rat livers, a very large increase in HHK activity occurs over a period of approximately 18 h (Figure 2) and tends to precede the proliferation of hepatocytes [4,10]. HHK activity was also induced in livers of rats fed on a choline-deficient ethionine-supplemented diet that is hepatotoxic, and, in this case, the HHK activity was found to be associated with the proliferating and differentiating liver progenitor (oval) cells . Elevated HHK activity was also found in human fetal liver and in hepatocellular carcinoma tumour tissue, whereas the HHK activity of the surrounding normal liver tissue was negligible . In addition, Tan et al.  found that HHK activity was also very high in a tumorigenic liver progenitor cell line (PIL-2) derived from a p53-knockout mouse. All of these findings led to the suggestion that HHK activity is an oncodevelopmental marker. In hepatocytes, liver progenitor cells, Walker-256 carcinosarcoma cells, P. polycephalum and thymus, the HHK activity was primarily located in the nucleus.
Time course of change in HHK activity (■) in liver nuclear extracts and hepatocyte proliferation (▲) in livers of rats following partial hepatectomy
None of the HHKs from animal cells has been characterized to any great extent, and, from our experience, purification and identification of HHK has proved very difficult. However, it has been found that HHK from regenerating liver phosphorylates histone H4 on both His18 and His75 to form 1-phosphohistidine, whereas that from Walker-256 carcinosarcoma cells forms 3-phosphohistidine . HHK from P. polycephalum only phosphorylates His75 to form 1-phosphohistidine . The HHK from S. cerevisiae has been purified to homogeneity  and partially characterized. The enzyme is a monomeric 32 kDa protein that catalyses the phosphorylation of His75 in histone H4, but not His18, to form 1-phosphohistidine and is specific for histone H4 in the histones. The S. cerevisiae HHK uses ATP as a substrate, with a Km value of 60 μM for ATP and 17 μM for histone H4 and has a requirement for a free divalent metal cation such as Mg2+or Co2+ . Genistein, a known tyrosine kinase inhibitor, has also been found to inhibit S. cerevisiae HHK with a Ki of 300 μM as a non-competitive inhibitor with respect to both ATP and histone H4 . Even though the S. cerevisiae HHK was purified, no identification of the enzyme has been published.
Histone H4 histidine phosphorylation and its biological role
Although in most of the work on HHKs, exogenous histone H4 was used as substrate for in vitro reactions, there is some evidence of phosphorylation of histone H4 in cells expressing high levels of HHK activity. Histone H4 isolated from regenerating livers of rats that had been injected with 2 mCi/100 g [32P]Pi after partial hepatectomy was analysed to reveal the presence of [32P]phosphohistidine [3,4]. We have also shown the presence of phosphohistidine in histone H4 isolated from PIL-2 cells . Chen et al.  showed that newly synthesized histone H4 was not a target for histidine phosphorylation in regenerating rat liver, and Huebner et al.  showed that histone H4 in nucleosome core particles was not phosphorylated in P. polycephalum.
His18 is in the highly basic N-terminal tail of histone H4 and normally interacts with an adjacent nucleosome via the acidic region of the of the H2A–H2B dimer . In addition, His75 is close to the DNA-binding site in the nucleosome core and forms a hydrogen bond with a glutamate residue of histone 2B, helping to stabilize the histone octamer  (Figure 3). On the basis of these interactions, Besant et al.  proposed that histone H4 phosphorylation may occur during DNA replication when histones are displaced from the DNA and inhibit the formation of histone–histone and histone–DNA interactions.
Part of the structure of the nucleosome complexed with 146 base pairs of DNA
Protein histidine phosphatases that act on phosphohistidines in histone H4
Kim et al. [18,19] showed that yeast PPs (protein phosphatases) 1, 2A and 2C (PP1, PP2A and PP2C), which are known serine/threonine phosphatases, also catalyse the dephosphorylation of histone H4 that had been phosphorylated on His75 by S. cerevisiae HHK. In this study, the ratios of the value of kcat/Km for histidine-phosphorylated histone H4 compared with that for the phosphoserine-containing substrate (phosphorylase a for PP1 and PP2A, myosin P-light chain for PP2C) were 11, 5 and 0.8 respectively . Thus histidine-phosphorylated histone H4 was found to be a better substrate for PP1 and PP2A than their normal phosphoserine-containing phosphoprotein substrates and was almost as good as the normal phosphoserine substrate for PP2C; however, the phosphohistone H4 was not a substrate for PP2B or the protein tyrosine phosphatase PTP-1 .
In 2002, a specific PHP (protein histidine phosphatase) was identified and characterized [20,21]. We decided to use chemically phosphorylated peptides based on the peptide sequences containing His18 or His75 from histone H4 as a way of examining the substrate specificity of PHP with respect to peptide sequence and phosphohistidine isoform . In this study, PHP was found to catalyse the dephosphorylation of both 1- and 3-phosphohistidine in both peptides, with the kcat/Km for 1-phosphohistidine being higher than that for 3-phosphohistidine by at most a factor of 2 . The peptide based on His75 was a poorer substrate than that based on His18 by a factor of 2–3, in terms of kcat/Km values. Histone H4 phosphorylated by a thymic HHK preparation did not appear to be a substrate of PHP, suggesting that structural factors may play a role in determining substrate specificity as well as amino acid sequence around the phosphohistidine residue .
Analysis of HHK activity and histone H4 phosphorylation
Measurement of HHK activity has largely relied on the differences in stability between the common phosphoamino acids to acid and alkali treatment. Phosphohistidine residues are acid-labile and alkali-stable, whereas phosphoserine and phosphothreonine are acid-stable and alakali-labile and phosphotyrosine is acid-stable and alkali-stable. Originally Wei et al.  developed a filter-based assay using 32P incorporation (from [γ-32P]ATP) into histone H4 to detect alkali-stable phosphorylation, but later Tan et al.  developed the assay to be more specific for histidine phosphorylation of histone H4 so that alkali-stable acidlabile phosphorylation was measured.
Normally these assays are accompanied by phosphoamino acid analysis of the phosphorylated histone H4 to confirm the presence of phosphohistidine. Phosphoamino acid analysis to detect phosphohistidine involves either partial hydrolysis of phosphohistone H4 by 3 M KOH at 105°C  or by overnight incubation with pronase E in ammonium bicarbonate  followed by thin layer electrophoretic  or thin layer chromatographic  separation of the of the amino and phosphoamino acids. The phosphoamino acids are detected by phosphoimaging the thin layer plate to detect the incorporated 32P and staining the plate with ninhydrin to detect the phosphoamino acid standards.
More recently, mass spectrometric methods have been developed to detect and analyse histone H4 histidine phosphorylation. Zu et al.  showed that it was possible to perform amino acid analysis by positive- and negative-ion mode electrospray-ionization MS and to analyse the sites of histidine phosphorylation in histone H4 by positive-ion mode electrospray-ionization MS/MS (tandem MS). Phosphohistone H4 phosphopepetide detection by MALDI–TOF (matrix-assisted laser-desorption ionization–time-of-flight)-MS was also demonstrated . 31P- and 1H-NMR are useful for the analysis of the isoforms of phosphohistidine present in histone H4  and histone H4 phosphopeptides , since both 31P and 1H signals of 1-phosphistidine and 3-phosphohistidine have different chemical shifts.
The most promising and most recent methodological development has been the production of antibodies that specifically recognize phosphohistidine residues in the context of the histone H4 amino acid sequence . Unsuccessful attempts to raise antibodies against phosphohistidine have been made in our laboratory and even using a non-hydrolysable phosphonopyrrole-alanine analogue of 3-phosphohistidine (Figure 1) as a hapten failed to produce antibodies that recognized phosphohistidine in phosphorylated histone H4. Recently, Kee et al.  raised polyclonal antibodies using a triazolylalanine phosphonate analogue of 3-phosphohistidine (Figure 1) incorporated into a synthetic decapeptide of the histone H4 amino acid sequence 14–23, in which His18 was replaced by the analogue. Kee et al.  clearly demonstrated that the antibodies not only recognized phosphohistidine in histone H4, but also recognized phosphistidine only in position 18 and not in position 75. This work hopefully heralds the development of a suite of antibodies that not only simply detect histidine-phosphorylated histone H4, but also specifically detect which histidine residue is phosphorylated and which form of phosphohistidine is present.
Despite the fact that histone H4 histidine phosphorylation was discovered nearly 40 years ago, the field has developed with painful slowness. No HHK has been fully characterized, and the biological role of histone H4 histidine phosphorylation remains a matter for conjecture. However, over the last decade, there have been a number of technological and methodological developments that should enhance the rate of progress in this area of research, and the development of anti-phosphohistidine antibodies should prove crucial to investigating the biological role of histidine phosphorylation of not only histone H4, but also protein histidine phosphorylation generally, in animal cells.
Signalling 2011: a Biochemical Society Centenary Celebration: A Biochemical Society Focused Meeting held at the University of Edinburgh, U.K., 8–10 June 2011. Organized and Edited by Nicholas Brindle (Leicester, U.K.), Simon Cook (The Babraham Institute, U.K.), Jeff McIlhinney (Oxford, U.K.), Simon Morley (University of Sussex, U.K.), Sandip Patel (University College London, U.K.), Susan Pyne (University of Strathclyde, U.K.), Colin Taylor (Cambridge, U.K.), Alan Wallace (AstraZeneca, U.K.) and Stephen Yarwood (Glasgow, U.K.).