Trifluoroethanol-induced conformational transition of the C-terminal sterile alpha motif (SAM) of human p73

https://doi.org/10.1016/j.abb.2017.02.004Get rights and content

Highlights

  • TFE-induced conformational transition of SAMp73 is complex.

  • The TFE-populated states does not appear during the folding of SAMp73.

  • TFE does not mimic a membrane-like environment for SAMp73.

  • Isolated peptides in TFE have similar conformations to that of SAMp73 in TFE.

Abstract

The alpha splice variant of p73 (p73α), a homologue of the tumour suppressor p53, has at its C terminus a sterile alpha motif (SAM); this domain, SAMp73, is involved in lipid binding and it is thought to mediate in protein-protein interactions. As SAMp73 is a 68-residue-long helical bundle, it could be a good model to study the (2,2,2-trifluoroethanol) TFE-induced conformational transitions of α-helical proteins. Furthermore, as SAMp73 binds to lipids through a well-known polypeptide patch, we can test whether TFE is a good mimic of lipids and membranes. To address those questions, we used several biophysical probes, namely, fluorescence, circular dichroism, 1D, 2D and 3D-NMR spectroscopies, and dynamic light scattering. The TFE-induced conformational transition of SAMp73 was complex, involving several species as detected by the biophysical probes. The last TFE-induced transition occurred at a concentration of TFE of ∼20% (v/v), where the protein lost its compactness. None of those TFE-induced species accumulated during the two-state folding of SAMp73 in aqueous solution. The final state at 40% TFE was highly helical, but its structure was not rigid. For SAMp73, TFE did not properly mimic a membrane-like environment, since at very low TFE concentrations, other residues, together with those known to interact with lipids, were also affected by the co-solvent. Comparison with studies on isolated peptides, comprising the helical regions of SAMp73, suggests that peptides were good models of the intact protein in TFE.

Introduction

Characterization of partially folded states of a protein is crucial in elucidating how a linear polypeptide chain folds itself into the unique three-dimensional structure. The 2,2,2-trifluoethanol (TFE) was the first alcohol used to allow accumulation of those partially folded states [1]. Under particular solution conditions (such as pH or solvent composition), the otherwise quite unstable partially folded species of a protein can be stabilized to be studied at equilibrium. Alcohols do not only induce partially folded states, but also they can mimic the conditions around membranes due to their physical-chemical features [2]. Although most of proteins are soluble, many of them must carry out their function in the membrane proximity, or alternatively, they must cross membranes to reach their cellular destination. Therefore, studies of proteins with alcohols in vitro may provide clues on their particular non-native species adopted: (a) in transport or degradation within the living cell; or, (b) in stress conditions caused by disease or environmental changes. From studies with large (>100 amino acids) folded proteins, it has been learnt that alcohols stabilise the secondary structure, but destabilize the tertiary one [2], [3], [4], [5].

The p73 protein (a member of the p53 family [6], [7], [8], [9]) contains in its sequence a transactivation domain (TA), a sequence-specific DNA binding polypeptide patch, and an oligomerization region. Due to this organization, p73 can take the place of p53 inducing apoptosis in tumour cells, even though its function in tumour suppression is not known [10]. However, both proteins differ in domain organization and gene structure. For example, the genomic structure of p73 at the 3’-end has three exons which encode several spliced forms (α to ζ) [11], [12]. These forms have different C-terminal extensions, with dissimilar expression patterns and functions among healthy tissues [12], [13].

The p73α has an extended C terminus consisting of a sterile alpha motif (SAM) and a C-terminal tail. The isolated SAM can modulate the function of p73 TA [14], although the particular region involved is unknown. In general, SAMs intervene in regulating protein function via self-association, association with other SAMs, lipids, or interactions with nucleic acids [15], [16], [17]. The structures of SAMs reveal a well-conserved fold of five α-helices, with small variants in the arrangement and the length of the helices [18, and references therein]. The structure of monomeric SAMp73 has been solved by NMR [19] and X-ray [20], [21]. The structure (Fig. 1) is formed by α-helix1, α1 (Pro491-Gly499); α-helix2, α2 (Ile506-Thr510); a 310-helix, α3 (Ile517-Gln521); α-helix 4, α4 (Ile525-Leu531); and α-helix 5, α5 (Arg538-Gln550) (in the numbering of the intact p73). We have described the conformational stability of SAMp73 at the residue level [22], [23]; moreover, we have shown that SAMp73 binds to lipids [24]. Furthermore, we have described the conformational features of the isolated α-helices in aqueous solution and in the presence of TFE [25].

In this work, we studied the conformational transitions of SAMp73 induced by TFE. Our purpose to start this study was three-fold. First, we aimed to decipher whether the most affected residues by the presence of TFE were the same whose environments changed by the presence of lipids [24]. Second, we were trying to find out whether there was a relationship between the structure acquired by protein fragments in the presence of TFE [25] and that of the intact protein in the same co-solvent. And finally, since SAMp73 is 68-residue-long, α-helical protein, we were trying to obtain clues about the behaviour of small α-helical proteins in the presence of TFE. To address those questions, we used several biophysical techniques: fluorescence, circular dichroism (CD), NMR and dynamic light scattering (DLS). Our results show that the tertiary structure of SAMp73 was firstly disrupted in the presence of TFE; concomitantly, there was an increase in protein size. Disruption of the secondary structure occurred at a higher TFE concentration. The TFE-induced conformational transition was also followed at residue-level by using NMR. However, when the tertiary structure was disrupted, the conformational exchange was intermediate-to-slow within the NMR time-scale, leading to broadening of most of the signals. The final highly helical state at 40% (v/v) TFE was devoid of any tertiary structure, as suggested by the lack of dispersion in the NMR spectra. In addition, its helical structure was not rigid, as shown by two pieces of evidence: (i) the lack of cooperativity (i.e., sigmoidal behaviour) in thermal denaturations; and, (ii) the absence of solvent-exchange protection.

Section snippets

Materials

Deuterium oxide and d3-TFE (96.5% purity) were obtained from Apollo Scientific (Stockport, UK). Sodium trimethylsilyl [2,2,3,3-2H4] propionate, TSP, and TFE (non-deuterated) were from Sigma (Barcelona, Spain). Standard suppliers were used for all other chemicals. Water was deionized and purified on a Millipore system. TFE concentrations will be reported from now on as % (v/v), and their solution properties will be calculated as described [26].

Protein expression and purification

SAMp73 was produced and purified in LB media as

The TFE-induced conformational transition, monitored by different spectroscopic probes, did not follow a simple two-state mechanism

As discussed in the following paragraphs, the TFE-induced transition of SAMp73 did not follow a two-state behaviour because there were different midpoints in the curves, obtained by the several techniques, with Eq. (1). However, use of more complicated equations (assuming three- and four-state transitions [41], [42]) to fit the experimental data obtained did not yield statistically significant different midpoints to those obtained from the use of the simplest equation (from a paired t-Student

The TFE-induced structural transition in SAMp73: possible implications for folding

The first conclusion of our studies is that TFE disrupted the tertiary structure of SAMp73, as it happens with other well-folded proteins [2]. However, our results with an all α-helical model protein indicate that the TFE-transition did not follow a simple two-state process, but rather, the structure around some tyrosine residues seemed to be lost before the secondary one (∼23% TFE) and the environment around the sole Trp542 (∼25%) (Fig. 3 A). At both transitions, there were changes in the

Funding sources

This work was supported by the Spanish Ministerio de Economia y Competitividad [CTQ 2015-64445-R to JLN and BIO 2016-78020-R to ACA]. The funding agency did not have any role in the research.

Acknowledgements

We thank Prof. Paul Fitzpatrick for handling the manuscript. We also thank the two reviewers for suggestions, discussion and ideas. JLN thanks Prof. C. H. Arrowsmith (Toronto, Canada) for the kindly gift of the SAMp73 clone. He also acknowledges May García, María del Carmen Fuster and Javier Casanova for excellent technical assistance.

References (56)

  • A. Kundu et al.

    Biophys. Chem.

    (2004)
  • G.I. Melino et al.

    Trend Biochem. Sci.

    (2003)
  • H.J. Lee et al.

    Structure

    (2012)
  • F.N. Barrera et al.

    J. Biol. Chem.

    (2003)
  • G. Bodenhausen et al.

    Chem. Phys. Lett.

    (1980)
  • D. Pantoja-Uceda

    J. Santoro. J. Magn. Reson

    (2008)
  • J. Sancho et al.

    J. Mol. Biol.

    (1992)
  • J. Sancho

    Arch. Biochem. Biophys

    (2013)
  • S.E. Jackson

    Fold. Des

    (1998)
  • R.W. Woody

    Methods Enzymol.

    (1995)
  • S.M. Kelly et al.

    Biochem. Biophys. Acta

    (2005)
  • M.P. Williamson

    Prog. Nuc. Mag. Spec.

    (2013)
  • M. Goodman et al.

    J. Am. Chem. Soc.

    (1962)
  • M. Buck

    Q. Rev. Biophys.

    (1998)
  • M. Perham et al.

    Biochemistry

    (2006)
  • K. Gast et al.

    Protein Sci.

    (1999)
  • F. Murray-Zmijewski et al.

    Cell Death Differ.

    (2006)
  • A. Yang et al.

    Nat. Rev. Mol. Cell. Biol.

    (2000)
  • K.H. Vousden et al.

    Nat. Rev. Mol. Cell. Biol.

    (2007)
  • A. Rufini et al.

    Genes Cancer

    (2011)
  • V. De Laurenzi et al.

    Cell Death Differ.

    (1999)
  • V. De Laurenzi et al.

    J. Exp. Med.

    (1998)
  • T.J. Grob et al.

    Cell Death Differ.

    (2001)
  • G. Liu et al.

    J. Biol. Chem.

    (2005)
  • J. Schultz et al.

    Protein Sci.

    (1997)
  • T. Avid et al.

    Nat. Struct. Biol.

    (2003)
  • T.M. Tanaka-Hall

    Nat. Struct. Biol.

    (2003)
  • S.W. Chi et al.

    EMBO J.

    (1999)
  • Cited by (3)

    • Tumor suppressor gene DLC1: Its modifications, interactive molecules, and potential prospects for clinical cancer application

      2021, International Journal of Biological Macromolecules
      Citation Excerpt :

      The domain exists in more than 200 human proteins, including enzymes, transcription factors, and signaling proteins. Most SAM domains are thought to be mainly involved in homologous or heterologous protein-protein interactions, although certain SAM domains have been reported to bind RNA and lipids [88–90]. The head-to-tail polymers of SAM can scaffold large macromolecular complexes [91,92].

    • Structural investigation of a C-terminal EphA2 receptor mutant: Does mutation affect the structure and interaction properties of the Sam domain?

      2017, Biochimica et Biophysica Acta - Proteins and Proteomics
      Citation Excerpt :

      Similarly, conformational studies of tendamistal, which represents an all β-protein, in presence of TFE also reveal a partially folded state with loss of tertiary structure but retention of the β-sheet core and gain of some unstable helical content in native loop regions [64]. Very recently the conformational transitions in the Sam domain of p73 (p73-Sam) were studied with several biophysical techniques in presence of increasing amount of TFE: at 20% TFE the protein globular compactness is lost whereas at 40% TFE the protein is characterized by elevated helical content but a relevant flexibility [60]. A structural dissection approach of p73-Sam and structural analysis of the resulting peptide fragments in presence of TFE indicate agreement with the conformational study on the intact protein and show that actually peptides can be considered as decent model-systems instead of entire proteins for such studies [65].

    View full text