The acquisition and storage of information in memory require specific long-lasting changes in gene expression. These changes have been proposed to depend upon chromatin remodelling, and on site-specific and dynamic post-translational modifications (PTMs) of histone proteins in brain cells. In the chromatin, histones and DNA are tightly associated and form nucleosomes. Each nucleosome contains a histone octamer composed of two heterodimers of the core histones H2A and H2B, and a tetramer of the core histones H3 and H4, around which 146 base pairs of DNA are wrapped 
. Nucleosomes are separated from each other by a short stretch of internucleosomal DNA bound to the linker histone H1. All core histones, their variants and linker H1 are known to be subjected to PTMs 
, which are covalent modifications that can occur on selective amino acids and that can be induced and erased by complexes of chromatin-modifying enzymes. Whilst over 200 histone PTMs have been identified in the brain 
, only few have been well characterised and shown to be linked to specific brain functions 
. The role/s of most, however, remain unknown.
Some of the best characterised PTMs on histones have been studied in the brain in relation to memory formation and include Lys acetylation, Ser/Thr phosphorylation, and Lys/Arg methylation. In particular, the level of phosphorylation on H3S10, acetylation on H3K9, H3K14, H4K5, H4K8 and methylation on residues including H3K4 and H3K9, have all been shown to be correlated with some forms of memory 
. These PTMs are established by an ensemble of enzymes comprising histone acetyltransferases (HATs), protein kinases and histone methyltransferases (HMTs), and are erased by histone deacetylases (HDACs), protein phosphatases and histone demethylases (HDMs) 
. Further to being induced directly by specific enzymes, histone PTMs are also subjected to multiple cis
regulatory cross-talk. This results in the establishment of specific combinations of PTMs thought to form a gene-specific ‘histone code’ that determines the level of transcriptional activity 
. The impact of histone PTMs on gene activity is, in part, mediated by specific reader and effector proteins that can bind in the presence (or absence) of specific PTMs. For instance, the HDM JMJD2A associates with chromatin only when H4K20 is methylated, but not when neighbouring sites are phosphorylated or acetylated 
. Selective interactions between neighbouring histones are also regulated by PTMs. Thus, residues 16–20 of H4 can interact with two acidic patches on the adjacent C-terminus of H2A, but this interaction is prevented by H4K16 acetylation by KAT8, leading to an increase in the local accessibility of the DNA to the transcriptional machinery 
Determining the ensemble of histone PTMs and identifying their different combinations and cross-talk are essential steps for the understanding of gene regulation. This is particularly relevant to the brain, because many brain functions are regulated by gene expression. Histone PTMs contribute to this dynamic regulation of gene expression, as they can alter the accessibility of DNA to the transcriptional machinery by opening or closing the chromatin 
. Over the past decade, great progress has been made in the identification and mapping of histone PTMs, and in the characterisation of the enzymes that catalyse them 
. Mass spectrometry (MS) has been particularly instrumental 
, and led to the detection of many PTMs on individual histones, and to the generation of comprehensive maps of histone PTMs in several species 
. However, a drawback of conventional ‘bottom-up’ MS methods is that proteins are typically digested into short peptides prior to MS. This generates complex biological samples that contain a mixture of short peptides coming from independent copies of the same proteins. This means that PTMs occurring simultaneously on a given histone cannot be determined because the peptides generated from this histone cannot be identified 
. Recently, however, this limitation has been circumvented by new MS techniques, specifically electron transfer dissociation (ETD) and electron capture dissociation 
. These techniques have allowed the analysis of long peptides (>20 aa), and have led to the analyses of PTMs co-occurring on individual histones 
. However despite these techniques, little progress has been made in the identification of PTMs in vivo
. Most studies to date have been carried out in cultured cells. They have therefore been limited by in vitro
conditions that often do not fully reflect the in vivo
situation, particularly in relation to the adult brain, where most neurons are postmitotic.
Here, we report on a novel approach that captures the status of histone PTMs and their combinatorial patterns directly in the adult brain in mice. This approach is based on ETD and collision induced dissociation (CID) high mass accuracy MS/MS using an Orbitrap XL-ETD. It allowed us to detect and identify multiple PTMs on individual histones in the adult mouse brain, and determine their combinations and association rules. Furthermore, this approach newly revealed the presence of atypical PTMs such as Ser and Thr acetylation, and Lys propionylation, butyrylation and crotonylation on specific histones, and of several novel motifs flanking phosphorylation, methylation and acetylation sites. The ensemble of these data provides important new insight into the histone code in the adult brain that may be relevant for complex brain functions.