Histone Serotonylation: Can the Brain Have “Happy” Chromatin?

Abstract

Recent work from Farrelly et al. (2019) indicates that histone tails can be serotonylated, suggesting a previously unappreciated direct mechanism of potential crosstalk between bioactive amines and the epigenome.

Since the discovery of histone acetylation and methylation in the 1960s, dozens of different covalent histone modifications including phosphorylation, ubiquitylation, propionylation, butyrylation, and crotonylation have been identified, affecting nearly every histone tail residue (Huang et al., 2014). While the activation or repression of various cell signaling and metabolic pathways often indirectly alters chromatin states, direct links between signaling molecules and chromatin are rare. Surprisingly, Farrelly et al. (2019) recently reported that the neurotransmitter, serotonin (5-HT) can be covalently attached to histone H3 by transglutaminase 2 (TGM2) (Figure 1). This finding raises the interesting question of whether alterations in neurotransmitter signaling associated with a myriad of emotions and behaviors might be recorded in a sort of chemical diary on the epigenome.

Figure 1. Histone Serotonylation by Transglutaminase 2.

High levels of nuclear serotonin (5-HT) lead to TGM2-dependent serotonylation of the histone H3 tail at glutamine 5 (H3Q5(5HT)), which co-occurs with lysine 4 trimethylation (H3K4me3). H3K4me3/H3Q5(5-HT) double-modified chromatin is found at promoters and correlates with gene transcription.

Although serotonin was discovered over 60 years ago (Berger et al., 2009), most studies of serotonin biology center around non-covalent interactions between serotonin and plasma membrane receptors resulting in the activation of signaling cascades primarily through GPCRs (Millan et al., 2008). Only in the last two decades have studies highlighted receptor-independent functions for serotonin and other monoamine neurotransmitters where transglutaminases act to covalently link free monoamine substrates to diverse protein targets (Muma and Mi, 2015; Walther et al., 2011). Colloquially known as “meat glue” in the food industry for their ability to crosslink proteins, transglutaminases catalyze a variety of chemical reactions including the hydrolysis or attachment of mono-amines to peptides through deamidation and monoaminylation, respectively (Lorand and Graham, 2003). Transglutaminases can also covalently link bioactive monoamines like serotonin, dopamine, and histamine to effector proteins such as small GTPases, and extracellular matrix components (Muma and Mi, 2015; Walther et al., 2011). Building on previous observations suggesting that transglutaminases modify histones (Ballestar et al., 1996), Farrelly et al. (2019) describe the direct incorporation of serotonin into chromatin during neural progenitor cell differentiation and hypothesize that histone serotonylation might regulate transcription.

Farrelly et al. (2019) go on to show that histone 3 glutamine 5 (H3Q5) serves as the primary site of serotonylation by TGM2 and that H3Q5 serotonylation (H3Q5(5-HT)) co-occurs with trimethylated H3 lysine 4 (H3K4me3), which correlates with active transcription. Although H3K4me3 and H3Q5(5-HT) occur concomitantly, TGM2-dependent serotonylation does not require H3K4me3, nor does the presence of H3Q5(5-HT) alter the efficiency of H3K4me3 methylation by the MLL1 complex. Consistent with a known role for H3K4me3 in recruiting RNAPII-dependent transcriptional machinery (Vermeulen et al., 2007), proteomics and co-immunoprecipitation studies suggest that H3K4me3/H3Q5(5-HT) doubly modified histones enhance the binding of TFIID to chromatin. This potentially increased recruitment of the basal transcriptional machinery raises the possibility that histone serotonylation might promote or potentiate gene expression. Histone modifications act in a combinatorial manner to regulate chromatin dynamics and gene expression in part via the recruitment of reader proteins. While H3K4me3 is recognized by the PHD domain of TAF3 (Vermeulen et al., 2007), which is a subunit of TFIID, whether there is a specific reader for H3Q5(5-HT) that contributes to the enhanced TFIID binding remains to be determined. Structural analysis of TFIID associated with serotonylated histones or peptides and further molecular studies would enhance our understanding of the mechanism that underlies the ability of H3K4me3/H3Q5(5-HT) to promote TFIID binding and transcription.

Further investigation into the molecular biology of chromatin monoaminylation may also reveal both upstream regulators and downstream consequences of TGM2 activity on histones. Beyond the requirement for high serotonin levels in the nucleus, the mechanisms controlling the recruitment of TGM2 to genomic loci and of TGM2-dependent catalysis of chromatin modifications remain largely unknown. The observation of histone serotonylation also invites speculation as to whether chromatin might be modified by other neurotransmitters or at additional residues. Transglutaminases can cross-link several monoamines, including dopamine, norepinephrine, and histamine, to proteins such as fiine, nore (Muma and Mi, 2015), suggesting that additional neurotransmitters might serve as substrates to modify chromatin. Earlier work using monodansylcadaverine and other substrates points to multiple glutamine residues on histone core proteins as potential transglutaminase targets in vitro (Ballestar et al., 1996). Additional studies may shed light on whether other histones and other H3 amino acid residues can be modified under physiological conditions.

In addition to now being in the position to address many exciting mechanistic questions governing the causes and consequences of histone monaminylation, it is equally exciting to consider the potential impact of these modifications on brain function. The long-term stability of a subset of covalent histone modifications such as methylation leads some researchers to hypothesize that altering chromatin modification patterns might provide a molecular basis for cognition, emotion, and other aspects of brain function via the perpetuation of gene expression patterns. However, our understanding of the mechanisms linking neural activity to changes in chromatin structure and function remains limited. Future studies may reveal whether histone serotonylation can act to stably record on the epigenome chronic changes in monoamine levels induced by events like learning or mood disorders, or if these marks serve as more transient mediators of acute signaling events. Notably, transglutaminases can remove monoamines through deamidation reactions (Lorand and Graham, 2003), and the deamidation of histones has been observed (Lindner et al., 1998), suggesting that histone monoaminylation is likely reversible. Beyond the brain, whether histone monaminylation is inherited by daughter cells following mitosis or by offspring from parents through intergenerational epigenetic inheritance remain open questions.

Given the many functions of serotonin signaling in different tissue contexts (Berger et. al 2009), the work by Farrelly et al. (2019) suggests a role for chromatin monoaminylation in regulating diverse aspects of normal physiology and development. Interestingly, Farrelly et al. showed that the H3K4me3/H3Q5(5-HT) antibody also detects signals in other brain cell types, like astrocytes, and in heart, colon, and blood samples. Therefore, histone monoaminylation may function in other contexts such as the gastrointestinal tract, which synthesizes and stores a majority of the total serotonin in the body (Berger et al., 2009), or in the pancreas where β cells require transglutaminase activity for optimal insulin secretion (Muma and Mi, 2015; Walther et al., 2011). It will also be interesting to consider whether monaminylation of chromatin may play any part in the etiology of neurotransmitter-related diseases (Berger et al., 2009). For instance, additional studies may uncover whether selective serotonin reuptake inhibitors (SSRIs), or other small molecules acting on monoamines, exert their effects through direct chromatin modification, and if these chromatin changes determine therapeutic efficacy.

The discovery of histone serotonylation reveals an exciting new possibility where small molecules involved in cell-cell signaling can be directly linked to chromatin and adds an additional level of complexity to both chromatin-regulatory and neurotransmitter-dependent signaling networks. We are excited for the many avenues of potential new research aimed at determining both the upstream and downstream biochemical mechanisms of his-tone serotonylation and the functional significance of this mark in both normal physiology and disease.