Research Article: Pupylated proteins are subject to broad proteasomal degradation specificity and differential depupylation

Date Published: April 22, 2019

Publisher: Public Library of Science

Author(s): Juerg Laederach, Hengjun Cui, Eilika Weber-Ban, Horacio Bach.


In actinobacteria, post-translational modification of proteins with prokaryotic ubiquitin-like protein Pup targets them for degradation by a bacterial proteasome assembly consisting of the 20S core particle (CP) and the mycobacterial proteasomal ATPase (Mpa). Modification of hundreds of cellular proteins with Pup at specific surface lysines is carried out by a single Pup-ligase (PafA, proteasome accessory factor A). Pupylated substrates are recruited to the degradative pathway by binding of Pup to the N-terminal coiled-coil domains of Mpa. Alternatively, pupylation can be reversed by the enzyme Dop (deamidase of Pup). Although pupylated substrates outcompete free Pup in proteasomal degradation, potential discrimination of the degradation complex between the various pupylated substrates has not been investigated. Here we show that Mpa binds stably to an open-gate variant of the proteasome (oCP) and associates with bona fide substrates with highly similar affinities. The proteasomal degradation of substrates differing in size, structure and assembly state was recorded in real-time, showing that the pupylated substrates are processed by the Mpa-oCP complex with comparable kinetic parameters. Furthermore, the members of a complex, pupylated proteome (pupylome) purified from Mycobacterium smegmatis are degraded evenly as followed by western blotting. In contrast, analysis of the depupylation behavior of several pupylome members suggests substrate-specific differences in enzymatic turnover, leading to the conclusion that largely indiscriminate degradation competes with differentiated depupylation to control the ultimate fate of pupylated substrates.

Partial Text

In mycobacteria and many other actinobacteria, covalent modification of proteins with the small (60–70 residues), intrinsically disordered protein Pup (prokaryotic ubiquitin-like protein) allows them to be recognized and degraded by a bacterial proteasome complex consisting of the 20S proteasome core particle and a ring-shaped ATPase referred to as Mpa (mycobacterial proteasomal ATPase) in mycobacteria or ARC (ATPase forming ring-shaped complexes) in other actinobacteria [1, 2]. Although this proteasomal degradation pathway is not essential under standard culture conditions, it provides bacteria with a critical advantage under certain stress conditions [3]. For example, the human pathogen Mycobacterium tuberculosis (Mtb) makes use of this pathway to persist inside host macrophages, while its non-pathogenic relative Mycobacterium smegmatis (Msm) gains advantage from the Pup-proteasome system under nitrogen starvation and DNA damage stress [4–8]. The modification of cellular proteins with Pup involves two structurally homologous and evolutionarily related enzymes, the Pup ligase PafA (proteasome accessory factor A) and the Pup deamidase/depupylase Dop (deamidase of Pup) [9–11]. The Pup ligase attaches the side-chain carboxylate of the C-terminal glutamate residue of Pup to a lysine side chain in the target protein by forming an isopeptide bond [12–15]. In mycobacteria, Pup is encoded with a C-terminal glutamine that first must be deamidated to glutamate by the enzyme Dop to produce the side-chain carboxylate for ligation [10]. The cleavage of the C-N bond in the glutamine side chain is chemically equivalent to the cleavage of the isopeptide bond of a pupylated protein. Hence it is not surprising that Dop also catalyzes the depupylation reaction [16–18]. Recognition of pupylated substrates by the proteasome complex occurs at the hexameric AAA+ Mpa-ring [19, 20], which is formed of three parts: one ring tier is made of the C-terminal AAA+ modules that stack on top of the 20S particle, followed N-terminally by a narrower collar-like double-ring tier formed by two consecutive β-barrel domains [21, 22]. From this tier emerge N-terminal helices that are pairwise engaged into a total of three coiled-coils. Upon binding to Mpa, Pup adopts a helical conformation in part of its sequence and associates via this helix with the coiled-coils of Mpa, forming a shared, three stranded coil [23, 24]. Pup’s N-terminus remains unstructured and available for threading into the Mpa central pore. Pup binds to Mpa with a 1:1 stoichiometry despite the presence of three coiled-coils [14], which is due to space constraints in the vicinity of the Mpa ring pore [20]. Mpa, similar to eukaryotic and archaeal proteasomal ATPases, uses a C-terminal interaction motif with a penultimate tyrosine to dock onto the 20S core particle. However, efficient degradation of pupylated substrates by the Mpa-proteasome in vitro could only be observed when an open-gate variant of the 20S core particle (oCP) is used, in which the N-termini of the α-subunits are truncated by seven residues [21, 25]. A recent X-ray structure of Mpa suggests that the reason might be found in the formation of a stable β-grasp domain at the very C-terminus of Mpa burying the C-terminal GQYL interaction motif inside the Mpa central pore [22]. It was suggested that association even with the oCP is strongly hindered by the formation of this domain. This, however, contradicts earlier studies that have shown stable interaction between Mpa and oCP in vitro using size exclusion chromatography or electron microscopy [20, 26].

In mycobacteria, the Pup-proteasome system has been shown to increase bacterial fitness under certain stress conditions, be it nitrogen starvation for M. smegmatis (Msm) or persistence in macrophages of M. tuberculosis (Mtb) [3–6]. Careful analysis of the structures and activities of the pupylation enzymes, namely the ligase PafA and the deamidase/depupylase Dop, has led to a good understanding of the reaction mechanisms by which pupylation and depupylation occur [9, 12, 13, 15–17, 40]. Furthermore, several studies have been conducted to identify pupylated substrates of different actinobacterial organisms, generating a catalog of pupylation targets [8, 27–34]. However, few key examples of Pup-modified proteins have been found, that upon degradation can increase the chance of survival under certain conditions [41, 42]. While for the substrates in question and a few others degradation has been shown, it is still unclear why many members of the pupylome show no difference in steady-state levels with impaired proteasomal degradation. For this reason, we investigated the possible discrimination between different pupylated proteins as degradation targets for the Mpa-proteasome complex.

Chemicals were purchased from Sigma-Aldrich and Media from Difco, if not otherwise stated.