Phosphorylated and Phosphomimicking Variants May Differ—A Case Study of 14-3-3 Protein

Abstract

Protein phosphorylation is a critical mechanism that biology uses to govern cellular processes. To study the impact of phosphorylation on protein properties, a fully and specifically phosphorylated sample is required although not always achievable. Commonly, this issue is overcome by installing phosphomimicking mutations at the desired site of phosphorylation. 14-3-3 proteins are regulatory protein hubs that interact with hundreds of phosphorylated proteins and modulate their structure and activity. 14-3-3 protein function relies on its dimeric nature, which is controlled by Ser58 phosphorylation. However, incomplete Ser58 phosphorylation has obstructed the detailed study of its effect so far. In the present study, we describe the full and specific phosphorylation of 14-3-3ζ protein at Ser58 and we compare its characteristics with phosphomimicking mutants that have been used in the past (S58E/D). Our results show that in case of the 14-3-3 proteins, phosphomimicking mutations are not a sufficient replacement for phosphorylation. At physiological concentrations of 14-3-3ζ protein, the dimer-monomer equilibrium of phosphorylated protein is much more shifted towards monomers than that of the phosphomimicking mutants. The oligomeric state also influences protein properties such as thermodynamic stability and hydrophobicity. Moreover, phosphorylation changes the localization of 14-3-3ζ in HeLa and U251 human cancer cells. In summary, our study highlights that phosphomimicking mutations may not faithfully represent the effects of phosphorylation on the protein structure and function and that their use should be justified by comparing to the genuinely phosphorylated counterpart.

Contemporary biophysical approaches for studying 14-3-3 protein-protein interactions

“…In isolation, each of the techniques FP, FRET, ITC, and SPR act as a unique lens through which to study 14-3-3 PPIs, but these methods alone do not provide a full picture. Other approaches not discussed here (e.g., microscale thermophoresis (Centorrino et al, 2022;Kozeleková et al, 2022), differential scanning fluorimetry (Cossar et al, 2021;Waløen et al, 2021;Joshi et al, 2022), analytical ultracentrifugation (Horvath et al, 2021;Leysen et al, 2021;Neves et al, 2021;Pohl et al, 2021;Srdanović et al, 2022), single angle x-ray scattering (Kast and Dominguez, 2019;Leysen et al, 2021) also contribute new understanding and support hypotheses. NMR (Kuusk et al, 2019;Neves et al, 2021) and native mass spectrometry (Bellamy-Carter et al, 2021) are also important techniques that shows great promise for studying 14-3-3 PPIs.…”

“…An analogous approach was also used to study how mutation of 14-3-3ζ Ser58 to Glu affected dimerisation. ( Kozeleková et al, 2022 ). The effect of shifting 14-3-3 dimerisation equilibria (when 14-3-3 protein is titrated from very low to very high concentrations) has been proposed as an explanation for some unexpected biphasic FP binding curves (see Section 2.2 ) ( Srdanović et al, 2022 ).…”

Autonomous Synthesis of Functional, Permanently Phosphorylated Proteins for Defining the Interactome of Monomeric 14-3-3ζ

“…The dimeric status of 14-3-3 isoforms ζ, ε, γ, η and β is regulated by phosphorylation of a serineSer58 in 14-3-3ζat the dimer interface, leading to its monomerization (Figure A). , Phosphorylated, monomeric 14-3-3 is thought to sensitize cells to apoptosis, sparking interest in the development of therapeutics that stabilize it . But studying this form of 14-3-3 has been challenging due to an inability to make it: Asp/Glu phosphomimetic mutations are reported to weaken the 14-3-3 dimer, albeit poorly compared to phosphorylation, ,,, and kinase phosphorylation results in incomplete modification due to the hidden interfacial location of this residue. − Characterizations of monomeric 14-3-3 have thus historically relied on deleting the entire N-terminal dimerization interface, adding interface mutations or, very recently, a multistep strategy to purify the phosphorylated away from the unmodified form. , Using pSer and PermaPhos GCE, we expressed and then purified homogeneous 14-3-3ζ with Ser, pSer, nhpSer, as well as Glu at position S58 (Figure B), and we evaluated their oligomeric status using SEC-MALS (Figure C and Table S2). The 14-3-3ζ WT eluted as a single dimeric peak at 50 μM initial concentration (∼2-fold above physiologic concentration), as did the phosphomimetic S58E variant, consistent with previous work. , In contrast, both 14-3-3ζ pSer58 and 14-3-3ζ nhpSer58 eluted exclusively as a monomer.…”

“…Phosphorylation at this site primes cells for apoptosis by monomerizing 14-3-3, − creating interest in anticancer therapeutics that act to stabilize the monomeric form . Efforts to study this phosphorylated form of 14-3-3 by direct enzymatic phosphorylation of the 14-3-3 dimer at this interfacial residue proved ineffective. − Prior work to understand the role of Ser58 phosphorylation therefore has largely focused on using phosphomimicking mutations (S → D/E), which unfortunately only partially monomerize 14-3-3, − as well as a variety of mutated forms of 14-3-3 that achieve monomerization by including multiple mutations (up to 7) at the dimer interface . It remains unclear how the monomerization of 14-3-3 by phosphorylation, as it occurs inside the cell, alters its interactome and ability to regulate client function in a way that primes cells for apoptosis.…”

“…In-principle, this limitation can be overcome by substituting phospho-sites with nonhydrolyzable (i.e., stable), functional analogs or mimics (Figure ). The negatively charged amino acids aspartate and glutamate (Figure A) are trivial to incorporate into proteins, but their charge, shape and hydrogen-bonding geometries are all markedly different than those of pSer and, in the case of 14-3-3 studies, they are insufficient to promote 14-3-3/client interactions ,, and also fail to fully monomerize 14-3-3. ,,, …”