Short N-terminal disordered regions and the proline-rich domain are major regulators of phase transitions for full-length UBQLN1, UBQLN2 and UBQLN4

Background:

UBQLN1, UBQLN2, and UBQLN4 are three related proteins that play essential roles in the regulation of protein homeostasis. These proteins are part of a family known as ubiquilins, which are involved in various cellular processes, including the ubiquitin-proteasome system and autophagy. These processes are critical for maintaining cell health and preventing the accumulation of misfolded or damaged proteins. UBQLN1 contains a ubiquitin-like (UBL) domain and a ubiquitin-associated (UBA) domain, which allows it to bind to ubiquitinated proteins and deliver them to the proteasome for degradation. UBQLN2 is another member of the ubiquilin family, and it shares similar structural features with UBQLN1, including the UBL and UBA domains, with the exception of the proline-rich (Pxx) region. Importantly, mutations in UBQLN2 have been linked to certain cases of amyotrophic lateral sclerosis, suggesting its involvement in the pathogenesis of this neurodegenerative disorder. The third member of the ubiquilin family is UBQLN4, which also shares structural similarities with UBQLN1 and UBQLN2. This family of proteins possesses highly conserved oligomerization C-terminal STI1-II domain and putative heat-shock protein binding domain STI1-I. Here, the authors of the preprint demonstrate different phase separation behavior among ubiquilins. Also, this study shows that different epitope tags used to study the protein in vivo or in vitro could potentially influence phase separation behavior.

Importance of the study:

How do different ubiquilin proteins carry out specific functions in the cellular environment, despite their highly identical, conserved domain architecture? Here the authors describe that distinct phase separation behavior among UBQLN1, UBQLN2, and UBQLN4, could potentially allow their engagement in different physiological conditions. Moreover, the authors stress the need for careful epitope tag design and additional in vitro tests to avoid experimental artifacts while studying tagged ubiquilins in vivo.

 Key findings:

UBQLN1, UBQLN2, and UBQLN4 proteins were shown to form liquid-like droplets in 200 mM NaCl. Although UBQLN1 and UBQLN2 exhibit 74% sequence identity, UBQLN2 and UBQLN4 exhibit 56% sequence identity. UBQLN2 and UBQLN4 droplets were larger than those of UBQLN1, indicating a higher phase separation propensity for UBQLN2 and UBQLN4. Also, the greater temperature dependence of UBQLN2 phase separation in comparison to UBQLN1 and UBQLN4 differentiates their condensation property.

An intrinsically disordered region (IDR) of ubiquilins – N-terminal to the UBL domain – is understudied and largely considered to be part of the UBL domain. For the first time, the authors could demonstrate that the deletion of the IDR made UBQLN1, UBQLN2, and UBQLN4 indistinguishable in terms of temperature-dependent phase separation behavior. Furthermore, the removal of the Pxx region from UBQLN2 abolished the sharp temperature dependence observed for full-length UBQLN2. Also, the addition of UBQLN2 Pxx into UBQLN4 modestly increased the temperature dependence of UBQLN4 phase separation behavior.  The authors were able to show a positive regulatory role of the UBQLN2 Pxx domain and a negative role of the IDRs in guiding phase separation of ubiquilins.

The authors hypothesized that the more negative charge of UBQLN1 and the neutral charge of UBQLN2 in IDR could be a cause of the low and high phase separation propensity, respectively. They then demonstrated that the triple variant G12R/D15R/E21Q (RRQ) of UBQLN1 mimicked the UBQLN2 phase separation propensity and that the triple variant R12G/R15D/Q21E (GDE) of UBQLN2 mimicked UBQLN1 phase separation propensity.

Figure 1 of Dao et al. (2023). Figure reproduced under a CC-BY-NC-ND 4.0 International License. A) Domain organization of UBQLN1, UBQLN2 and UBQLN4. B) Temperature-concentration phase diagrams for the ubiquilins by measuring saturation concentrations (c_sat) at different temperatures indicated that UBQLN2 and UBQLN4 more efficiently phase separate than UBQLN1.

Figure 3 of Dao et al. (2023). Figure reproduced under a CC-BY-NC-ND 4.0 International License. A) Sequence alignment of N-terminal IDRs of ubiquilins showed a more negative or neutral charge for marked residues in UBQLN1, and a more positive or neutral charge for marked residues in UBQLN2. B) The phase diagram showed more negative or positive charge residue replacement in UBQLN2 or UBQLN1, respectively switching their phase separation behavior.

Often epitope tagging is critical for the in vitro or in vivo study of a given protein. Considering the sensitive phase separation behavior of ubiquilins, the authors stress that careful selection of epitope tags is very crucial. Widely used tags like FLAG and Myc could provide a higher negative charge to the protein. To avoid that, a charge-neutral ALFA tag could be potentially used to study ubiquilins. Furthermore, the presence of aromatic amino acids in the tag could improve phase separation behavior. For example, the HA tag possesses three tyrosine amino acids and has been shown to substantially increase phase separation. The authors could show that a tyrosine-free, charge-neutral tag perturbs ubiquilin phase separation the least.

Figure 4 of Dao et al. (2023). Figure reproduced under a CC-BY-NC-ND 4.0 International License. A) Epitope tag sequences and net charges (tyrosine residues in cyan). B) Phase diagram of untagged or tagged UBQLN2, indicates that lower negative charge and higher tyrosine residues in tag positively influenced phase separation behavior. C)  Phase diagram of untagged or tagged UBQLN4, indicates that lower negative charge and higher tyrosine residues in tag positively influenced phase separation behavior.

References:

Boeynaems S, Alberti S, Fawzi NL, Mittag T, Polymenidou M, Rousseau F, Schymkowitz J, Shorter J, Wolozin B, Van Den Bosch L, Tompa P, Fuxreiter M. Protein Phase Separation: A New Phase in Cell Biology. Trends Cell Biol. 2018 Jun;28(6):420-435. doi: 10.1016/j.tcb.2018.02.004. Epub 2018 Mar 27. PMID: 29602697.

Dao TP, Kolaitis RM, Kim HJ, O’Donovan K, Martyniak B, Colicino E, Hehnly H, Taylor JP, Castañeda CA. Ubiquitin Modulates Liquid-Liquid Phase Separation of UBQLN2 via Disruption of Multivalent Interactions. Mol Cell. 2018 Mar 15;69(6):965-978.e6. doi: 10.1016/j.molcel.2018.02.004. PMID: 29526694.

Gerson JE, Linton H, Xing J, Sutter AB, Kakos FS, Ryou J, Liggans N, Sharkey LM, Safren N, Paulson HL, Ivanova MI. Shared and divergent phase separation and aggregation properties of brain-expressed ubiquilins. Sci Rep. 2021 Jan 11;11(1):287. doi: 10.1038/s41598-020-78775-4. PMID: 33431932.

Jin X, Zhou M, Chen S, Li D, Cao X, Liu B. Effects of pH alterations on stress- and aging-induced protein phase separation. Cell Mol Life Sci. 2022 Jun 24;79(7):380. doi: 10.1007/s00018-022-04393-0. PMID: 35750966.

Riley JF, Fioramonti PJ, Rusnock AK, Hehnly H, Castañeda CA. ALS-linked mutations impair UBQLN2 stress-induced biomolecular condensate assembly in cells. J Neurochem. 2021 Oct;159(1):145-155. doi: 10.1111/jnc.15453. PMID: 34129687.

Tags: epitope tags, n-terminal disordered region, phase separation, ubiquilins

doi: https://doi.org/10.1242/prelights.35795

Read preprint (No Ratings Yet)