br Conflicts of interest br Financial support

Conflicts of interest
Financial support NZ is an Established Investigator of the Dutch Heart Foundation (2013T111) and is supported by an ERC Consolidator grant (617376) from the European Research Council and by a Vici grant from the Netherlands Organization for Scientific Research (NWO; 91818643). AL is supported by a Dekker grant from the Dutch Heart Foundation (2016T015). JT is supported by an AMC PhD fellowship.
Author contributions
Acknowledgments
Introduction Ubiquitination is a well-characterized post-translational modification that is critical for regulating a wide range of cellular processes in eukaryotic organisms, including higher plants [1], [2], [3]. The conjugation of ubiquitin (Ub) to target proteins is sequentially carried out by E1 Ub-activating enzymes, E2 Ub-conjugating enzymes, and E3 Ub-ligases [4], [5], [6]. In plant genomes, there are few E1s, approximately 40 E2s, and more than 1000 E3s [7], [8]. With their abundance and target-binding activities, E3s generally determine the specificity of the ubiquitination pathway. In humans, however, E2s not only interact with E1s and E3s to receive and transfer Ub, respectively, but also regulate, at least in part, the length and topology of the poly-Ub chain and the efficiency of poly-ubiquitination conducted by E3s [5], [9]. Thus, E2s are not simply involved in ubiquitination, but are one of the regulators in the ubiquitination pathway. E3s contain one of three distinct functional domains: HECT, RING, or U-box [10], [11], [12]. The U-box motif shares a similar structure with the RING domain, but does not bind zinc ions to act as an E3 Ub-ligase [11], [12]. Compared to humans and yeast, higher plants have a large number of U-box motif-containing E3 Ub-ligases. Arabidopsis, a dicot model plant, contains 64 U-box genes [13] and the monocot model crop rice possesses at least 77 U-box genes [14]. Based on their primary sequences and presence of specific domains, the U-box E3s are divided into nine different Beta-Lapachone mg [14]. Class II and III U-box E3s are the most abundant E3s and are typified by the presence of a protein–protein interacting armadillo (ARM) repeat domain [13], [14], [15]. Plant U-box proteins have a role in diverse plant-specific phenomena, including hormone signaling [16], [17], responses to abiotic/biotic stress [18], [19], [20], self-incompatibility [21], and flowering time control [22]. For instance, rice ARM-U-box E3 SPL11 negatively regulates programmed cell death [20]. SPL11 plays an additional role in regulating flowering time by mono-ubiquitinating its target protein SPIN1 that represses flowering by down-regulating the flowering promoter gene HD3A[22]. OsPUB15, which encodes a class II ARM-U-box E3, regulates oxidative stress and cell death responses by reducing reactive oxygen species in rice [19]. To function as Ub-ligases, E3 proteins must interact with E2s. As compared to the extensive studies on E3s, functional studies on E2s are relatively rudimentary in higher plants. We previously reported that there are 48 genes encoding Ub-conjugating (UBC) fold-containing putative E2 proteins in the rice genome that are divided into three classes [23]. In another study, Kraft et al. [24] sorted 37 Arabidopsis E2s into 14 groups (groups III–XVI) based on detailed sequence homology analyses. In addition, Arabidopsis has a single SUMO-conjugating enzyme (AtSCE1a) and two Related to Ub-conjugating (RUB) enzymes (RCE1 and RCE2) that were classified into groups I and II, respectively [24]. In this report, we further classified the 48 rice E2s into 15 different groups. Yeast two-hybrid (Y2H) assays were performed to examine the interaction profiles of 40 E2s with 17 ARM-U-box E3s in rice. Of the 40 E2s, 11 E2s belonging to groups VI, VII, and VIII accounted for 70% of the E2–E3 interactions. These E2–E3 interactions were further validated by in vitro self-ubiquitination assays with rice SPL11 ARM-U-box E3 and various E2 partners. SPL11 E3 displayed distinct self-ubiquitination patterns, including poly-ubiquitination, mono-ubiquitination, and no ubiquitination, depending on the various E2s. Overall, these results suggest that the mode of ubiquitination of SPL11 E3 is critically influenced by individual E2s.