1. Introduction

During ubiquitination, isopeptide bonds are formed between the Ub carboxyl terminus and the ε-amino group of lysine (K) residues within the target protein, via sequential reactions catalyzed by E1 activating enzymes, E2 conjugating enzymes, and E3 ligases. One Ub could be modified by another Ub on K6, K11, K27, K29, K33, K48, K63, and M1 residues to form various types of Ub chains. The modifications by different type or length of these linkage Ub chains can confer different functions or fate to the modified protein, such as leading the modified protein to degradation, modulating signaling pathways, and regulating enzyme activity [2]. Modification of a protein by a K48-linked polyubiquitin chain generally targets the protein to the 26S proteasome for degradation [3], and while the cellular consequences of modification by K6-, K11-, K27-, K29-,or K33-linked polyubiquitin chains are not well understood, some studies have suggested that these modifications may also participate in the 26S proteasome-mediated degradation of target proteins [4]. Conversely, monoubiquitination and K63-linked polyubiquitination are widely known to regulate intracellular processes such as DNA damage repair, genomic stability, protein activity, inflammation, apoptosis, endocytic trafficking, and translation [3]. Besides Ub, eukaryotes also encode many ubiquitin-like (UbL) modifiers, such as SUMO-1,-2 and -3, NEDD8, FAT10, ISG15, UFM1, Hub1, etc., which share a higher structure similarity to Ub but have distinctive amino acid sequence and perform different functions from Ub [5]. Ub or UbL modifications are reversible through isopeptide bond proteases, deconjugating these modifiers from target proteins. The enzymes deconjugating UbL modifiers rarely share cross-reactivity with deubiquitinase (DUB). The rigor of these modifications is conducive to the meticulous regulation on the function of cellular proteins and increases the complexity of intracellular regulation.

Some proteases possess a catalytic triad, commonly Ser/His/Asp in serine proteases or Cys/His/Asp in cysteine proteases. This catalytic triad generates a nucleophile with which to attack the carbonyl group of a target peptide, and thus enables hydrolysis. The residue that is critical in this nucleophilic attack varies depending on the type of protease, with serine being essential in serine proteases and cysteine being essential in cysteine proteases. In order to generate a nucleophile, however, all that is required is the presence of a nucleophilic group such as a sulfhydryl or hydroxyl, that can be polarized and activated by the histidine residue of the catalytic triad [6]. In human, approximately 100 proteases for deconjugating Ub or Ub chains were found and classified into two main families, cysteine proteases and metalloproteases [7]. The cysteine proteases include six subfamilies, ubiquitin-specific proteases (USPs), ubiquitin C-terminal hydrolases (UCHs), ovarian tumor proteases (OTU), and Machado–Josephin domain proteases (MJDs), while metalloproteases only comprise the Jab1/Mov34/Mpr1 Pad1 N-terminal+ (MPN+) (JAMM) domain proteases, that are zinc-dependent metalloproteinases [8]. Cysteine protease family usually contains a canonical catalytical diad consisting of Cys and His, or triad, including Cys, His, and Asp residues [9]. Despite the fact that varied deubiquitinating enzymes are folded into different catalytic domains, the cysteine residues are almost at the same position to catalyze the deconjugation of the isopeptide bond [9]. These DUBs with distinct spatial structures and catalytic properties accurately regulate diverse cellular processes. The cystine residue, C90 in hUSP1 or C91 in chUSP1, are the nucleophilic group and indispensable for deubiquitinating activity of USP1. The histidine residue in the catalytic core, such as H594 in chUSP1 or H594 in hUSP1, uses the positive charge of their sidechain at neutral pH to promote enzyme-substrate interactions, and these histidine residues are indispensable for deubiquitinating activity of USP1 as well [10-12].

2. Data

DUBs participate in chromatin regulation [13], virus infection [14], tumorigenicity [15], and immune regulation [16]. In particular, human USP1 (hUSP1) is critical for the regulation of DNA damage repair and genomic stability [17,18]. Furthermore, by associating with human USP associated factor 1 (hUAF1), the deubiquitination activity of hUSP1 can be enhanced up to 35-fold [18]. The hUSP1/hUAF1 dimer has been shown to deubiquitinate monoubiquitinated FANCD2 and PCNA, thus modulating DNA damage repair and genomic stability [17,18], and to regulate the degradation of T cell receptors and CD4⁺ molecules on the surface of T lymphocytes upon viral infection, leading to their functional impairment [19]. USP1 is a member of the largest DUB superfamily, and is essential for DNA damage repair in human and in chicken cells [18,20]. The action and expression of hUSP1 depend on the phase of cell cycle, and synchronize with other proteins related to DNA synthesis or damage repair, such as Rad51 and PCNA. hUSP1, either in mRNA level or in protein level, is cell-cycled regulated, and approach the maximum during S phase [21]. DNA damage suppresses its expression or induces its degradation [18]. Furthermore, hUSP1 has been shown to help to maintain stem cell characteristics in both osteosarcoma cells and mesenchymal stem cells by deubiquitinating and thus stabilizing inhibitors of DNA binding, which in turn antagonize the basic helix-loop-helix transcription factors responsible for differentiation [22]. Both full-length and C-terminally truncated USP1 isoforms are able to deubiquitinate PCNA-Ub and FANCD2-Ub, and to regulate the DNA damage repair response and hence genomic stability [18,23]. However, autocleavage of hUSP1 in vivo could result in down-regulation and subsequently degradation of hUSP1 activity. This is critical for regulating hUSP1 activity during DNA damage monitoring and repairing [23,24]. Autocleavage sites mutated version of human USP1 were used to investigate its function in vivo or in vitro [23,25].

Besides hUSP1, hUAF1 stimulates hUSP12 and hUSP46 activity as well. The crystallography analysis shows that the charged surface on hUAF1 β-propeller is the key for the interaction between hUAF1 and hUSP46 finger subdomain [26]. Their catalytic triad (Cys, His, Asp) locates in palm subdomain of three USPs. The hUAF1 stimulation on deubiquitinase activity of USP46 could be eliminated by hydrophobic mutation of residues involved in their interaction [26]. hUSP1/hUAF1 shares similar mode of interaction with USP46/UAF1 complex [26]. hUAF1 also can interact with finger domain of hUSP12 to activate hUSP12 [27]. The activated complex of hUSP12 or hUSP46 by binding to hUAF1 could be activated further by binding to WDR20, another WD repeat protein, or binding to second hUAF1. This hyper-activation is not observed for the hUSP1/hUAF1 complex [28,29]. In addition, hUSP1 (785 residues) is larger than hUSP12 and hUSP46 (370 and 366 residues, respectively). hUSP1 is 31% identical to hUSP12 and hUSP46, although the latter two proteins share 88% identity. The crystallography analysis of hUSP1/hUAF1 complex has not been done yet. Binding of hUAF1 stabilizes hUSP1 and promotes interaction with PCNA-Ub and FANCD2-Ub [30], and the USP1/UAF1 complex promotes DNA double-strand break repair via homologous recombination [20]. Indeed, knocking out murine UAF1 causes a defect in homologous recombination and early embryonic lethality [31]. UAF1 and the USP1/UAF1 complex are also important in viral infection and pathogenesis, particularly in viral genomic integration and replication. The hUSP1/hUAF1 complex interacts with the human papilloma virus E1 DNA helicase to promote virus replication, and elimination of hUSP1/hUAF1 deubiquitinase activity dramatically impacted viral replication [32], and hUAF1 interacts with the herpesvirus saimiri Tip protein, promoting T cell receptor downregulation [33].

Although chUAF1 shares 98% identity with hUAF1, the homology of USP1 is only 72% between the two species. ChUAF1 could bind to chUSP1, both when they were co-expressed and as purified proteins, and that this binding enhanced chUSP1 activity as much as 54-fold [10], much higher than the reported 35-fold enhancement in human USP1 full-length [12,17,18] and 16-fold in truncated human USP1 that missing N-terminal 20 amino acids (hUSP1ΔN) [25]. It was observed that k_cat of hUSP1 and hUSP1ΔN activity were increased respectively 18.6-fold [18] and 7-fold [25] in the presence of hUAF1, while the k_cat of chUSP1^FL was increased 21-fold in current study. K_M of chUSP1^FL decreased 2.87-fold in the presence of chUAF1, while K_M of hUSP1 decreased 1.5-fold [12] or 2-fold [17,18]. These may suggested that the activation of USP1 upon binding to UAF1 may be due to an increase in k_cat with no drastic change in the K_{M ,} that was similar to what were observed on activation of human USP12 and USP46 binding to hUAF1 [25].

The ubiquitin regulation system is one of the most important regulating systems and the target of many pathogens in cells. Pathogens have evolved a series of molecular strategies to promote infection, proliferation, and survival during host–pathogen interactions. Viruses also encode some viral types of modifiers, relevant catalytic enzymes, or regulatory proteins to mimic the counterparts of the host cell [34]. Many viruses encode viral DUBs to hijack host cell defense mechanisms, such as Herpes simplex virus (HSV), Kaposi’s sarcoma-associated herpesvirus (KSHV), Epstein–Barr virus (EBV), MDV, pseudorabies virus (PRV), Human cytomegalovirus (HCMV), severe acute respiratory syndrome-coronavirus (SARS-CoV), and Middle East respiratory syndrome-related coronavirus (MERS-CoV) which encode USP type DUBs [35-37]. Equine arteritis virus (EAV), porcine reproductive and respiratory syndrome virus (PRRSV), Crimean–Congo hemorrhagic fever virus (CCHFV), and Dugbe virus (DUGV) encode OTU type DUBs [38,39]. Regardless of the fact that various viral-derived DUBs regulate the same intracellular target protein or a pathway as their host cell counterpart, they share no structural and sequence homology [34,40]. These differences allow the virus to break through the defense of its host, or avoid the removal of the virus by host cells. Some of these viral enzymes were used as targets of anti-virus drugs. There are virus-encoded DUBs that have been used to develop antiviral drugs, such as the two DUBs encoded by SARS-CoV and MERS-CoV. The inhibitors of the two enzymes can specifically resist the two viruses, respectively, and there is no cross-reactivity [41]. The activity of HIV-encoded protease, HIV-1, is critical for the packaging and infectivity of HIV [42]. HIV-1 enzymes have been targeted for the screening of anti-HIV drugs, among which specific inhibitors, saquinavir and ritonavir, have been approved to be used in clinical trials [43,44].

MDV-encoded DUB, UL36, is highly conserved in different virulent MDVs, indicating that the activity of UL36 is evolutionarily important to the pathogenicity and replication of MDV, and the UL36 catalytic domain could be used as a target to develop the drug for curbing the increasing virulence and pathogenicity of MDV [45]. MDV-encoded UL36 exists in MDV-induced T lymphoma cells for a long term [46], and chicken cells did not wipe off this foreign enzyme that seriously affects the intracellular environment promoting MDV activity. Some animal or pathogen DUBs, such as chicken USP1 [10], human USP21 [47], Legionella pneumophila SdeA [48], Plasmodium falciparum PfUCH54 [49], Epstein–Barr virus BPLF1 [50], reportedly exhibit a de-NEDDylating activity and can de-conjugate NEDD8-modified protein. MDV-encoded UL36 has no cross-reactivity on the NEDD8 substrate [45]. MDV-encoded UL36 enzyme possess a strict de-conjugating activity on Ub substrates but not cross-reactivity on UbL modifiers, UL36 is not able to hydrolyze SUMO1, 2 and 3, NEDD8, UFM1, FAT10, and ISG15 substrates [45]. MDV-encoded UL36 prefers to hydrolyze K11, K48, and K63 linkage Ub chains, and could partially hydrolyze K6, K29, and K33 Ub chains, but there was no activity on K27 Ub chains [45]. The two residues, C98 and H234, are essential for UL36-DUB activity, and D232 contributes to UL36 deubiquitinating activity but is not indispensable [45]. The efficiency of UL36-DUBs in hydrolyzing K48 and K63 types of Ub chains was higher than that on the linear type of Ub chains (M1) [45]. MDV could reverse the ubiquitylome of chicken T-cell lymphoma by comparison with normal CD4⁺ T cells, and the N-terminal 75 kDa fragment of UL36, which contains the viral DUB catalytic domain (UL36-DUB), exists in MDV-induced T lymphoma cells in high level [46]. UL36 is of Cys protease activity, Roche CPIC and serine protease inhibitor,such as phenylmethanesulfonyl fluoride (PMSF), fail to inhibit UL36-DUB activity [45].

The properties elucidated in these researches could promote further investigations for the insights in advancing the basic understanding of various DUBs’ intracellular functions including the genomic integration, tumorigenesis, and immunosuppression of some viruses. These findings have also laid the foundation for future screening of anti-virus drugs with viral DUB as a target.