**1. Introduction**

### *1.1. Overview of MYSM1 Protein Structure and Catalytic Activity*

In recent years, Myb-like, SWIRM, and MPN domains 1 (MYSM1) has emerged as an essential regulator of hematopoiesis, immunity, and other aspects of mammalian physiology. It is primarily a nuclear chromatin-interacting protein, with orthologues found only in the vertebrate species, indicating more recent evolutionary origins and suggesting specialized biological functions.

Structurally MYSM1 comprises SANT, SWIRM, and MPN domains (Figure 1). The SANT domain of MYSM1 is structurally similar to the DNA-binding domain of transcription factor cMYB [1,2]. It can bind to DNA in vitro [2] and is required for MYSM1 association with histones in vivo [3], however, whether the DNA-binding is sequence specific is not known. The SWIRM domain of MYSM1, in contrast, does not have direct DNA binding activity [4], and although it is a common domain-type

for chromatin-associated proteins [4,5], it was shown to be dispensable for MYSM1 interactions with histones, at least in some systems [3]. Importantly, the MPN metalloprotease domain of MYSM1 is the catalytic domain, characterized by Zn2<sup>+</sup>-binding, the JAMM motif with consensus sequence EXnHSHX7SX2D [6], and deubiquitinase catalytic activity (DUB) [6].

**Figure 1.** MYSM1 Protein: Domain Structure and Mutations in Human Patients. MYSM1 comprises SANT, SWIRM, and MPN domains. MYSM1 mutations reported in human MYSM1-deficiency syndrome patients include the p.H656R substitution within the critical JAMM-motif of the catalytic domain, and the nonsense variants p.E390\* and p.Arg478\* truncating MYSM1 protein upstream of the catalytic domain. All mutations in the patients are in a homozygous state.

Our knowledge of the physiological substrates of MYSM1 DUB-catalytic activity continues to expand. Histone H2A, monoubiquitinated at K119, was the first substrate of MYSM1 to be described (H2AK119ub) [7]. More recently MYSM1 was also demonstrated to cleave K63-polyubiquitin chains on cytosolic substrates TRAF3, TRAF6, and RIP2 [3,8]. MYSM1 catalytic activity against polyubiquitin chains of different geometries [9] was subsequently tested using in vitro assays, and MYSM1 was shown to cleave M1, K6, and K27 chains, but was catalytically inactive against K11, K29, K33, and K48 chains [8]. This suggests that our knowledge of MYSM1 substrates remains incomplete and further targets of its catalytic activity may be discovered in the future.

Despite some recent advances, the precise role of MYSM1 in shaping the epigenetic landscape of different mammalian cells remains poorly understood. Histone H2AK119ub is a repressive epigenetic mark deposited on chromatin primarily by polycomb repressive complex 1 (PRC1), for long-term gene silencing during cell differentiation and lineage specification [10–12]. In contrast, histone H2BK120ub is an epigenetic mark of transcriptionally active genes [13,14]. The catalytic activity of MYSM1 against histone H2AK119ub, but not H2BK120ub, therefore indicates its primary role as an activator of gene expression [7]. However, MYSM1 is one of many DUBs in mammalian cells that can catalyze H2AK119ub deubiquitination [15]. Other major DUBs with specificity for histone H2AK119ub over H2BK120ub are BAP1 and USP16, while USP3, USP12, USP22, and USP44 deubiquitinate both H2AK119ub and H2BK120ub, as well as different non-histone substrates [15]. How MYSM1 cooperates with the other DUBs to regulate the genome-wide landscape of histone H2A ubiquitination and the gene expression profiles of different mammalian cell types remains poorly understood. Furthermore, MYSM1 activities against polyubiquitinated forms of histone H2A at DNA damage foci [15,16] or against other less well-characterized ubiquitinated histones and histone variants [11,17,18] to our knowledge have not been investigated.

### *1.2. MYSM1-Deficiency in Human and Mouse: Mechanistic Insights and Biomedical Significance*

Characterization of *Mysm1*-deficient mouse strains has been instrumental in revealing the essential functions of MYSM1 in different aspects of mammalian physiology [19,20]. *Mysm1*-/- mice exhibit partial embryonic lethality, growth retardation, skeletal and coat pigmentation defects, and complex hematopoietic and immune phenotypes [19], while no phenotypic abnormalities are reported in *Mysm1*+/- heterozygous mice. The International Mouse Phenotyping Consortium (IMPC) [21–23] provides extensive primary data on these phenotypes [24]. This and the far-reaching data from recent publications will be covered in depth in subsequent sections of this review.

*MYSM1* mutations in patients with an inherited bone marrow failure syndrome (IBMFS) highlight the biomedical importance of understanding MYSM1 activities and functions. Five patients with homozygous *MYSM1* mutations were characterized in depth [25–27], carrying either a p.H656R substitution within the critical JAMM-motif of the catalytic domain, or a nonsense variant p.E390\* truncating MYSM1 protein upstream of the catalytic domain [25–27] (Figure 1). The variants were therefore predicted to severely impact or fully abolish MYSM1 catalytic activity. All five patients exhibit anemia and leukopenia [25–27], in some cases associated with growth retardation [26,27], developmental malformations [27], and neurodevelopmental delay [27]. B cells were severely depleted in all patients [25–27], correlating with reduced serum IgM levels [25–27], although IgG levels were reduced in only one of the patients [26]. Most patients also showed a reduction in NK cell numbers, while neutropenia was noted in three patients, and T cell depletion was observed in two patients [25–27]. Skeletal and craniofacial abnormalities reported in two patients included limb shortening (rhizomelia) and midface hypoplasia [27]. Additionally, a sixth patient also carrying the p.E390\* *MYSM1* mutation in a homozygous state was reported to have neutrophilic panniculitis, as well as reduced B cell count, anemia, and a mild growth retardation [28]. Finally, a novel homozygous mutation p.R478\* in *MYSM1* was recently identified via whole-exome sequencing in a patient diagnosed with Diamond-Blackfan anemia, a disorder characterized by anemia and to a lesser extent other hematological and developmental abnormalities [29].

Multilineage defects in hematopoiesis in the *MYSM1*-deficient patients sugges<sup>t</sup> impaired hematopoietic stem cell (HSC) function, and this was further supported by several lines of evidence. Thus, a reduction in the frequency of CD34+ hematopoietic stem and progenitor cells was reported in the patient homozygous for the *MYSM1*:c.1967A>G:p.H656R mutation [26]. Remarkably, this patient experienced a spontaneous genetic reversion, restoring the normal sequence of one *MYSM1*-allele in hematopoietic cells and resulting in a correction of all immunohematological defects [26]. Assuming that the reversion mutation originated in a single HSC, we can conclude that restoration of MYSM1 function provided a very strong selective advantage and allowed one HSC clone to reconstitute normal wild-type hematopoiesis in competition with a large pool of MYSM1-deficient stem and progenitor cells. This attests to the true importance of MYSM1 for normal HSC function in human [26]. Consistent with this, immunohematological defects in several other *MYSM1*-syndrome patients were successfully cured via allogeneic hematopoietic stem cell transplantation (HSCT) [27].

The high MYSM1 protein homology between human and mouse (87%) and the similarities in the phenotypes of *MYSM1*-deficiency between the species indicate that studies of MYSM1 activities and functions in mouse models can provide important insights into the etiology of *MYSM1*-deficiency syndrome in human. The emerging data from many murine studies published over the past 8 years will be reviewed below.
