Intrinsic Disorder to Order Transitions in the ... - ACS Publications

Feb 22, 2016 - the respiratory syncytial virus (RSV) relies on a RNA polymerase complex with a phosphoprotein cofactor P as the structural scaffold, w...
0 downloads 0 Views 6MB Size
Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Article

"Intrinsic disorder to order transitions in the scaffold phosphoprotein P from the respiratory syncytial virus RNA-polymerase complex" Maria Gabriela Noval, Sebastián Andrés Esperante, Ivana Giselle Molina, Lucia B. Chemes, and Gonzalo de Prat-Gay Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.5b01332 • Publication Date (Web): 22 Feb 2016 Downloaded from http://pubs.acs.org on February 23, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

“Intrinsic disorder to order transitions in the scaffold phosphoprotein P from the respiratory syncytial virus RNApolymerase complex”†



This work was supported by grants: PICT-2011-0721 from ANPCyT; CONICET (Consejo Nacional de Investigaciones Científicas y Técnicas) PIP 2011-2014 and Ciência Sem Fronteiras (Rio de Janeiro, Brazil) Program Grants.

MGN and IGM hold a graduate fellowship from CONICET; SAE, LBC and GdPG are CONICET Career Investigators

María G. Noval1, Sebastian A. Esperante1, Ivana G. Molina1, Lucía B. Chemes1 and Gonzalo de Prat-Gay1,2.

1

Protein Structure-Function and Engineering Laboratory, Fundación Instituto Leloir and IIBBACONICET, Av. Patricias Argentinas 435, 1405 Buenos Aires, Argentina.

2

Ciência Sem Fronteiras Senior Fellow, CNPq, Laboratório de Genômica Estrutural, Instituto de

Biofísica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brazil.

To whom correspondence should be addressed: Gonzalo de Prat-Gay, Protein Structure-Function and Engineering Laboratory, Fundación Instituto Leloir, IIBBA-Conicet, Patricias Argentinas 435, (1405), Buenos Aires, Argentina. Tel: (0054) 1152387500 Fax: (0054) 1152387501 E-mail: [email protected]

1 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

Abbreviations: RSV, Respiratory Syncytial Virus; IDP, Intrinsically Disordered Proteins; CD,

Circular Dichroism; SEC, Size Exclusion Chromatography; SLS, Static Light Scattering; PII, Polyproline helix type-II; Gdm.Cl, Guanidinium Chloride; TFE, 2,2,2,-trifluoroethanol; ANS, 8anilino-1-naphthalenesulfonate.

2 ACS Paragon Plus Environment

Page 2 of 41

Page 3 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

ABSTRACT. Intrinsic disorder is at the center of biochemical regulation and is particularly overrepresented among the often multifunctional viral proteins. Replication and transcription of the respiratory syncytial virus (RSV) relies on a RNA polymerase complex with a phosphoprotein cofactor P as the structural scaffold, which consists of a four-helix bundle tetramerization domain flanked by two domains predicted to be intrinsically disordered. Since intrinsic disorder cannot be reduced to a defined atomic structure, we tackled the experimental dissection of the disorder-order transitions of P by a domain fragmentation approach. P remains as a tetramer at temperatures above 70 °C, but shows a pronounced reversible secondary structure transition between 10 and 60 °C. While the N-terminal module behaves as a random coil-like IDP independent of tetramerization, the isolated C-terminal module displays a cooperative and reversible metastable transition. When linked to the tetramerization domain, the C-terminal module becomes markedly more structured and stable, with strong ANS binding. Therefore, the tertiary structure in the Cterminal module is not compact, conferring “late” molten globule-like IDP properties, stabilized by interactions favored by tetramerization. The presence of a folded structure highly sensitive to temperature, reversibly and almost instantly formed and broken, suggests a temperature sensing activity. The marginal stability allows for exposure of protein binding sites, offering a thermodynamic and kinetic fine tuning in order-disorder transitions, essential for the assembly and function of the RSV RNA polymerase complex.

3 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 41

Intrinsically disordered proteins (IDPs) are functional proteins which, in the absence of their biological partners, lack unique and defined secondary or tertiary structure under physiological conditions of pH and ionic strength

1, 2.

This structural category differs from random coil

conformation due to the existence of discrete dynamic conformational ensembles of multiple structures within local regions, that can undergo structural transitions depending on the physicochemical environment 3-5. IDPs are involved in processes that depend on protein-protein, protein-nucleic acid and protein-small ligand interactions. In addition, IDPs are associated with many biological processes such as transcription, translation and replication, cellular signaling and cell cycle regulation, among others

6, 7.

IDPs may exist in at least three separate functional

conformations: molten globule-like IDPs, pre molten globule-like IDPs or random coil-like IDPs, depending on the environmental and structural context

2, 3, 6.

IDPs are abundant in nature, being

present in the archaea, bacteria and eukarya kingdoms8. A correlation exists between organism complexity and the intrinsic disorder content of their proteomes, suggesting a role in higher order regulatory and signaling functions. On the other hand, the proportion of IDPs is particularly high in viral proteomes 9, reaching up to 50% within RNA viruses with small genomes

10.

This

increased proportion of disordered regions has been associated with the high adaptability, mutation rates, and increased structural flexibility present in viruses, and is thought to allow interactions with multiple cellular and viral targets, conferring viral IDPs with their multifunctional nature 11, 12. The Mononegavirales order are negative single-stranded RNA viruses and include several human pathogens such as Ebola, Rabies, Mumps and Respiratory Syncytial Virus (RSV), among others 13. Viruses from this order have a viral RNA-polymerase complex (RPC), which is part of the virion, and consists of the nucleoprotein (N) wrapping the RNA forming a helical nucleocapsid (NC), the viral RNA-dependent RNA polymerase (L) and the polymerase cofactor phosphoprotein (P), which serves as an essential tethering factor between the different RPC components. Interestingly, within the family Paramyxoviridae subfamily Pneumovirinae, an additional protein is part of the RPC: the transcriptional anti-terminator M2-1 14. Notably, many N and P proteins from paramyxovirus RPCs have been identified as IDPs by computational methods 15 and experimental techniques such as gel filtration, light scattering, circular dichroism and nuclear magnetic resonance, among others

16, 17, 18.

The P protein is essential for viral

transcription and replication 19. It mediates the recognition of the nucleocapsid by the polymerase 4 ACS Paragon Plus Environment

Page 5 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

L and acts as a chaperone of N, mediating the stabilization of newly synthesized N protein (N0) by blocking non specific binding to RNA 19. P presents a conserved modular organization within the Paramyxovidae and Rhabdoviridae families, consisting of a central oligomerization domain forming either dimers, trimers or tetramers depending on the virus alternated order and disorder regions

11, 25, 26.

20, 21,22-24

that is flanked by

The RSV P phosphoprotein is the smallest protein

among their paramyxovirus counterparts and is found as a tetramer in solution

27,28.

Previous

experiments demonstrated that P displays a modular organization, containing a central trypsinresistant module predicted as coiled-coil (residues 104-160), which comprises the core tetramerization domain (residues 119-160). In turn, this domain is flanked by N-terminal (residues 1-103) and C-terminal (residues 161-241) modules that have been predicted to be intrinsically disordered

27, 29, 30,

although no experimental demonstrations have been reported

(Figure 1A). P presents several phosphorylation sites, with different turnovers

31-33.

Although it

has been described that phosphorylation is dispensable for genome replication 34, recent evidence indicates that P phosphorylation is involved in several key functions within the viral life cycle 35, 36.

Chemical stability studies showed that P exhibits two unfolding transitions 28, a first transition

occurring at low denaturant concentration (0-3.5 M Gdm.Cl) that does not involve tetramer dissociation, and a second transition corresponding to the unfolding and dissociation of the stable tetramerization domain (4-6 M Gdm.Cl)

28.

The first unfolding transition displays low

cooperativity but involves a substantial loss of α-helix content and disruption of hydrophobic patches, as judged by reduction of ANS binding capacity. This indicates that Gdm.Cl at low concentrations causes a decrease in side chain packing and a loss of fluctuating tertiary structure within P 28. Most of the interaction sites within P are encompassed in the N- and C-terminal regions, although very little is known about their conformational properties

37-41.

Particularly, based on

modeling and mutagenesis studies the N0 binding site (residues 1-29) 38 in the N-terminal region as well as the L-binding site (residues 212-239)

40

in the C-terminal region of P have been

proposed to constitute α-helical molecular recognition elements (α-MoREs). Conversely, for the NC-P interaction, it has been shown that only the last two residues of the interaction site located at the C-terminus of P (residues 231-241)

39, 41

become structured upon interaction, while

residues 231-239 remain flexible within the complex 39. Interestingly, two temperature sensitive mutations within the C-terminal module (G172S and E176G) affect the virus capacity to 5 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 41

replicate at physiological or restrictive temperatures (37 or 39 ºC, respectively) 42, 43 by disrupting the P-N interactions 42 Since RSV is the causative agent of acute bronchiolitis and lower tract respiratory diseases in children and newborns worldwide, and due to the lack of a vaccine or effective antivirals, RSV infection is considered an unsolved clinical problem

14, 44.

In this context, structural and

mechanistic understanding of the RPC assembly may allow the discovery of novel targets for antiviral design

45.

In spite of some high resolution structures for isolated RPC proteins or

domains being solved for paramyxoviruses and other Mononegavirales

19,

several of the RPC

components have not been amenable to crystallography studies or NMR, due to the presence of dynamic structural regions with different degrees of order-disorder transitions

15, 17, 18.

Considering the lack of structural information about RSV P, we to investigated the structural and conformational transitions of the full length protein and its module. For this purpose we tackled a fragmentation approach in combination with circular dichroism, fluorescence spectroscopy, and size exclusion chromatography as simple but sensitive and powerful techniques in the study of IDPs

46, 47, 48.

We provide the first experimental evidence that both the N- and C-terminal

modules of P have an IDP nature, but with different degrees of disorder. The N-terminal module presents characteristics of random coil-like IDP, with a conformational behavior independent of the tetrameric context and local secondary structural elements compatible with the presence of binding motifs. Conversely, the C-terminal module presents a metastable component rich in αhelix and modulated by temperature. The tertiary structure in the C-terminal module is not compact, conferring molten globule-like IDP properties, stabilized by interactions favored by tetramerization. The presence of a structure sensitive to temperature, reversibly and quickly formed and broken, suggest the existence of a temperature sensing activity, in line with reported temperature sensitive replicative mutations of P

42, 43.

RSV P can be considered as a core or

scaffold protein around which all the others RPC components are assembled, and we discuss the advantages of conformational flexibility modulating the interaction with target proteins favoring the regulation of viral transcription and replication.

6 ACS Paragon Plus Environment

Page 7 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

EXPERIMENTAL PROCEDURES. Plasmid constructions. The Human RSV strain-A2 P sequence (UniProtKB accession-number: P03421) was cloned between the BamH I/EcoR I sites of the pRSET-A vector (Invitrogen) where the 6xHis-tagged sequence has been removed by digestion with Nde I and BamH I restriction enzymes. The resulting plasmid was sequenced and used as template for site direct mutagenesis and a thrombin site (LVPRGS) was added within the P sequence to generate the different fragments by thrombin digestion of the recombinant P protein. For the different constructions thrombin cleavage sites were added between PN (residues 1 to 102) and PTETC (residues 103 to 241) or between PNTET (residues 1 to 161) and PC (residues 162 to 241), respectively (Figure 1A). All cloning was confirmed by sequencing. Protein Expression and purification. P phosphoprotein and PTETC, PN, PNTET, PC fragments. Protein expression of the different constructs were performed as previously described for the full length P 28. For protein purification the cell pellets were resuspended in 50 ml of lysis buffer (50 mM Tris.Cl pH 8.0, 600 mM NaCl, 2 mM 2-mercaptoethanol and 1 mM EDTA) per liter of culture, lysed by sonication, and supernatants were precipitated by adding solid ammonium sulphate to 40% w/v during 1 h at 4°C. The precipitated protein was collected by centrifugation, resuspended in 30 ml of 20 mM Tris.Cl pH 8.0 and 20 mM NaCl and dialyzed against 1 liter of the same buffer. After dialysis, the protein sample was incubated 25 minutes at 75 °C, and ultracentrifuged at 40000 rpm for 40 min at 4 °C. The resulting soluble fraction was treated with 1 mg of Ribonuclease A (Sigma) and 72U of DNAse-I (Sigma) incubated for 4 h at 37°C. Sample was subjected to ionic exchange Q-hyperD (Pall) in buffer 20 mM Tris.Cl pH 8.0, 50 mM NaCl and buffer with 1M NaCl was used to elute the phosphoprotein and the other thrombin variants. For P phosphoprotein, the sample was concentrated using Amicon centrifugal filter units (Millipore) and subjected to a size exclusion chromatography (SEC) on a Superdex 200 gel filtration column (GE, Healthcare). For the thrombin variants, samples were concentrated using Amicon and subjected to thrombin proteolytic digestion (1U/mg) in buffer Tris.Cl pH 7.5 with 50 mM NaCl and 2.5 mM CaCl2 for 4 h at 37 °C. The reaction was stopped by adding 1 mM of phenylmethylsulfonylfluoride (PMSF) followed by a Supperdex-200 gel filtration. All

chromatographic separations on

Superdex-200 were performed in 20 mM Tris.Cl pH 8.0 and 200 mM NaCl. The different fragments eluted from this column were >95% pure as judge by SDS-PAGE. Trypsin digested PTET and PC* fragments. The tryptic PTET fragment was obtained as previously described 28. PC* fragment. Purified PC

7 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

was digested with trypsin from bovine pancreas TPCK treated (Sigma-Aldrich) in 100 mM Tris.Cl pH 8.0, 50 mM NaCl and 50 mM CaCl2 for 45 minutes at 37 °C at a ratio 7500:1 (protein:trypsin w/w). The reaction was stopped by adding 1 mM PMSF. The digestion product was purified by reverse phase chromatography using a C8 HPLC column (Restek). The protein concentration was determined by quantitation of peptide bounds at 220 nm with a standard calibration curve. Sequence, molecular weight, purity and oligomerization state of P and the different fragments are detailed in Supplementary Figure 1 and Supplementary Table 1.

Determination of protein molecular weight and oligomerization state. MALDi-TOF measurements. The molecular weight of the different proteins or fragments was confirmed by MALDI-TOF mass spectrometry (Bruker Daltonics, Billerica, MA, USA). MALDI-TOF spectrometers measure m/z (mass/charge) ratios corresponding to mono- or multi-charged monomeric protein chains, as non-covalent interactions in oligomers that are not covalently bonded are usually disrupted during ionization. Therefore, monomeric molecular weight (and not native MW) is determined through this technique. The obtained values are shown in Supplementary Table 1. Static Light Scattering measurements. Protein oligomerization state was determined by static light scattering (SLS) using a Precision Detector PD2010 light scattering instrument connected in tandem to a high-performance liquid chromatography system and an LKB 2142 differential refractometer. The 90° light scattering intensity and refractive index signals of the eluting material were recorded on a PC computer and analyzed with the Discovery32 software supplied by Precision Detectors. The protein concentration of the samples was used in each SEC run to determine the average molecular weight was ~1 mg/ml. The general formula used by the equipment to estimate molecular weight is:

[Equation 1] Is MW = A⋅c ⋅(dn /dc)2 ⋅Io

Where MW is the sample molecular weight, Is and Io are the intensities of incident and radiated light, dn/dc is the change in refractive index with concentration, A is an instrumental constant and c is the concentration (in g/ml)

64.

The oligomeric state of each protein species was then

determined by dividing the MW estimated through SLS by the theoretical molecular weight 8 ACS Paragon Plus Environment

Page 8 of 41

Page 9 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

obtained

from

the

aminoacidic

sequence

using

the

ProtParam-Tool

(http://web.expasy.org/protparam/). Hydrodynamic radius analysis. SEC of all fragments was carried out on an analytical Superdex S200 HR 10/300 (24 ml) (GE Healthcare) column calibrated with a gel calibration kit (Pharmacia Biotech, Uppsala, Sweden). The void volume (Vo) and total volume (VT) were determined by loading Blue Dextran and acetone, respectively. All measurements were performed in an AKTA-FPLC equipment (GE Healthcare) at 0.5 ml/min flow rate. The buffers used in the runs were 50 mM Tris.Cl pH 7.5, 200 mM NaCl, or 50 mM sodium formate pH 3.0, 200 mM NaCl. Hydrodynamic behavior analysis. In order to analyze the hydrodynamic properties of the different P fragments, we first calculated the apparent molecular weight using the elution volumes obtained by SEC for each fragment (MWSEC). We used the calibration curve elution volumes (Ve) obtained from the globular standards to estimate partition coefficients for each standard as Kav=(Ve-V0)/(VT-V0). The plot of Kav versus log(MW) for the standards was then used as a calibration curve to estimate the MW of each protein fragment using linear regression and interpolation. Next, we created a normalized measure to compare the hydrodynamic properties of fragments of different molecular weight, which we called the “Molecular weight ratio” (MWSEC/MWSLS) where MWSLS is the native molecular weight determined by SLS and corresponds to the native oligomeric molecular weight of each fragment listed in Supplementary Table 1. This ratio represents the deviation from the hydrodynamic behavior expected for a globular protein (Ratio MWSEC/MWSLS =1). Chemicals and solutions. Unless stated otherwise, all measurements were performed in 20 mM Tris.Cl pH 7.5 buffer, 50 mM NaCl or in 20 mM Sodium formate pH 3.0 buffer, 50 mM NaCl. All chemical reagents were of analytical grade (Sigma Aldrich) and solutions were prepared with Milli-Q plus water and filtered through 0.22 µm membranes prior to use. Far-UV circular dichroism spectroscopy (CD). Far-UV CD measurements were carried out on a Jasco J-810 spectro polarimeter (Jasco, Japan) with cell paths of 0.1 and 0.2 cm.

The

temperature was set using a Peltier temperature-controller. CD spectra were recorded at standard sensitivity, 50 nm/min scanning speed, 8 s response time, 0.1 nm data pitch and a 4 nm bandwidth. All spectra were an average of at least 4 scans. Protein or fragments concentration

9 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 41

range between 5 and 40 µM depending on the experiment and fragment size. Raw data were converted to molar ellipticity, using the following equation 46:

[Equation 2]

Where deg is the raw CD signal in deg, [c] is protein concentration in Molar units, #bonds is the number of peptide bonds, and L is the path length in cm. Thermal denaturation experiments by Far-UV CD spectroscopy. Ellipticity at 220 nm was recorded as temperature was increased or decreased at a scan rate of 5 °C/min from 5 to 85 °C. Acquisition parameters were: 4 nm bandwidth, 4 s response time and 0.2 °C data pitch. Data analysis. The change in the CD signal during the thermal denaturation can be described by the following equation:

[Equation 3]

Where U0 and N0 are the mean residue ellipticity of the unfolded and native states, respectively. Um and Nm account for the linear variation of the signals with temperature. ∆G is the free energy of unfolding, R is the gas constant expressed as 1.9872 calK-1mol-1 and T is the temperature in °K. Since there are no molecularity changes during the thermal transition (Figure 4A), we fit our data to the Gibbs-Helmholtz equation that describes the folding transition of a monomer protein as a function of temperature 49: [Equation 4]

Where ∆G is the free energy of unfolding, ∆H is the enthalpy of unfolding, ∆Cp is the heat capacity of unfolding, Tm is the midpoint of the unfolding transition and T is temperature in Kelvin. We fit our thermal denaturation data to the combination of equations 3 and 4 49. Since ∆Cp is difficult to estimate from CD measurements, we used this model to estimate the Tm by 10 ACS Paragon Plus Environment

Page 11 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

setting ∆Cp to zero, which has been proposed as a reasonable assumption for initial calculations of thermal transitions that do not involve molecularity changes 49. This approach is validated by the fact that no molecularity changes are observed in the thermal transitions studied (P, PNTET and PTETC remain as tetramers and PN and PC as monomers throughout the transition). All reported Tm values and corresponding errors were obtained as the average of three to four independent experiments. We note that the PC fragment denaturation curve lacked a stable baseline, which decreases the reliability of the Tm estimated by data fitting. Therefore, these values were included in text only for the purpose of comparing these values to those of PTETC and full length P, as visual inspection clearly indicates a decrease in thermal stability. Gdm.Cl Chemical denaturation experiments. We performed the full unfolding transition (0-7.5M Gdm.Cl) analysis in separate tubes at 10 µM protein concentration, over night incubation and at 20 ºC as previously described 28. For the first unfolding transition (0-4M Gdm.Cl) titrations were performed at 20 ºC or 10 ºC with an incubation time of 5 minutes between each point, time where the system reached equilibrium (Figure 6 B). Alpha helical stabilization in 2,2,2-trifluoroethanol (TFE). TFE titration curves were carried out at 25 °C in 20 mM Tris.Cl pH7.5, 50 mM NaCl buffer containing varying concentrations of TFE (v/v)

46.

Data analysis. For the estimation of the number of residues stabilized in α-helix

conformation from the CD spectra experiments, we analyzed the value of the molar ellipticity at 220 nm at 0% and 50% TFE or at different temperatures and pHs by using the following empirical relationship to define the molar ellipticity value expected for a fragment with all its residues sampling an α-helical conformation (100% of α-helix) 46: [Equation 5]

Where n is the number of residues for a given fragment. We calculated the fraction of α-helix for each fragment in each conditions, and multiplying the fraction of helix by the number of residues we obtained the number of residues in α-helical conformation. We defined the parameter “αhelix propensity” as the difference between the number of residues calculated in stabilizing

11 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 41

conditions (low temperature, low pH, or high TFE) and the number of residues in buffer conditions (Supplementary Table 2). Unfolding kinetics. Thermal and chemical unfolding kinetics were followed by the change in the far UV-CD signal at 220 nm in 20 mM Tris.Cl pH 7.5, 50 mM NaCl buffer. Conformational changes triggered by temperature were performed doing temperature jumps between 10 to 40 ºC or vice versa. P phosphoprotein from a concentrated stocks held at 10 or 40 ºC were transferred to buffer equilibrated at 40 or 10 ºC in the CD cuvette, respectively.

Chemical unfolding

kinetics experiments were performed transferring P phosphoprotein from a concentrated stock held at 10 ºC to a cuvette containing buffer with 3.5 M Gdm.Cl or 6 M Gdm.Cl, respectively. The slow kinetic transition observed for the 6 M unfolding experiment was fitted to a simple exponential equation.

[Equation 6]

Where A is the amplitude of the transition, t is the time in minutes and k is the unfolding rate constant. For both experiments, we monitored the buffer baseline for 2 minutes, and performed the mixture of the protein with the buffer without stopping the trace. Fluorescence spectroscopy analysis. 8-anilino-1-naphthalenesulfonate (ANS) fluorescence emission spectra were recorded on a Jasco FP-6500 spectrofluorometer with an excitation wavelength of 370 nm and a scan speed of 100 nm/min, using excitation and emission bandwidths of 5 nm at 20 °C. Protein and fragments at different concentrations were incubated 20 minutes with 100 µM ANS, and the ANS binding capacity was evaluated by recording the fluorescence emission spectra from 400 to 600 nm. We analyzed the Gdm.Cl ability to disrupt the protein or fragment ANS binding capacity, by titrations with increased concentrations of Gdm.Cl and following the decrease of intensity of the fluorescence emission 475 nm spectral band.

12 ACS Paragon Plus Environment

Page 13 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

RESULTS. Conformational properties of the human RSV P phosphoprotein modules. As we have previously stated, the P phosphoprotein presents a modular organization composed of three modules: the N-terminal, the coiled-coil tetrameric- and C-terminal domains. To study the conformational properties of the different P modules we generated recombinant human RSVA2 full length phosphoprotein P and different fragments that include the monomeric N- and Cterminal modules (PN and PC), a shorter C-terminal fragment of PC (PC*) and the tetrameric variants comprising the isolated tetramerization domain (PTET), and the tetramerization domain plus the N- or C-terminal modules (PNTET and PTETC) (Figure 1A and Supplementary Table 1). In the present work, we will refer to PN and PC in the context of the full phosphoprotein P as “Nterminal module” and “C-terminal module”, respectively. Determinants of the anomalous hydrodynamic behavior of P. The P phosphoprotein displays a largely anomalous hydrodynamic volume

28, 29

but the determinants of this behavior cannot be

easily extracted from prediction algorithms. We used size-exclusion chromatography (SEC) 47 to evaluate the hydrodynamic behavior of the different P fragments (Supplementary Figure 1). In order to compare the hydrodynamic properties of fragments of different size and oligomerization state, we calculated the apparent molecular weight from SEC data (MWSEC) and normalized it to the native molecular weight (MWSLS) of each species by using the ratio MWSEC/MWSLS (see Experimental Procedures). The ratio MWSEC/MWSLS represents the deviation from the hydrodynamic behavior expected for a globular, ideally spherical protein (Ratio MWSEC/MWSLS =1). All fragments presented MWSEC/MWSLS ratios ranging from 2.7 to 3.9, consistent with an extended conformation (Figure 1B). The monomeric PN and PC fragments presented ratios (MWSEC/MWSLS) of 2.7 ± 0.1 and 3.1 ± 0.1, which can be explained by the extended conformation and low compactness of the polypeptide chain typical of IDPs. The anomalous behavior of the four helix bundle PTET (MWSEC/MWSLS) of 2.7 can be explained by the rigid and asymmetric extended nature expected for a coiled-coil structure

29.

The bi-modular fragments

PNTET and PTETC, and the full length P protein presented the highest deviations (MWSEC/MWSLS of 3.8 ± 0.3, 3.7 ± 0.1 and 3.9 ± 0.2, respectively), which may be explained by a combination of both disordered nature and coiled-coil conformations, indicating that the three P modules contribute to the largely anomalous hydrodynamic behavior described for P. 13 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 41

Secondary structure analysis and temperature-induced conformational changes. As a first approach to analyzing the secondary structure of individual and combined modules in comparison with full-length P, we measured far-UV CD spectra. The CD spectrum of P at 25 °C (Figure 2A) showed characteristics of α-helix, i.e., minima at 220 and 208 nm with the latter shifted to 206 nm which, together with a 220/208nm ratio ~0.76 (ratios of 1 are typical of pure αhelix) indicated the presence of a significant amount of disordered structure, as previously discussed

28.

Interestingly, the secondary structure content of P was largely dependent on

temperature variations (Figure 2A) with a substantial amount of structure remaining even at 85 °C (Figure 2A, gray line), in agreement with the complex thermal behavior suggested by Llorente

29.

However, in our hands this thermal behavior was fully reversible (Supplementary

Figure 2). We observed a structural transition between 5 and 37 °C (Figure 2A, black lines), that suggested the presence of a metastable structural element, in accordance to our previous observations for P chemical denaturation 28. The tetrameric fragment PNTET at 25 °C presented a CD spectrum similar to the full length protein (Figure 2B), suggesting that even with 30% less amino acid residues, the relative disorder/order proportion of PNTET is similar to that of P. However, no temperature transition was observed between 5 and 37 °C (Figure 2B, black lines). The monomeric PN fragment displayed a CD spectrum at 25 °C typical of a disordered conformation with a predominant minimum ~ 200 nm, no components of regular secondary structure, and the expected increase in secondary structure at high temperature due to strengthening of hydrophobic interactions, a typical feature of IDP regions (Figure 2C)

50.

In

contrast to PNTET, the CD spectrum at 25 °C of the tetrameric variant containing the C-terminal module, PTETC, presented a high proportion α-helical content, with a ratio 220/208 nm of 0.85 similar to that of the fully helical PTET domain (ratio 220/208 nm=1) (Figures 2D and 2E). Moreover, the conformational transition observed between 5 °C and 37 °C in full length P was also observed in PTETC, and the changes in the CD spectrum clearly indicated that the transition involved the loss of α-helical structure (Figure 2D). The monomeric PC fragment presented a CD spectrum characterized by a small negative band at 220 nm indicative of some degree of α-helix content and a minimum centered at 204 nm, which indicates a mixture of contributions from disorder (200 nm) and α-helix (208 nm), showing that PC features both structured and disordered elements in contrast to the fully disordered conformation of the PN fragment (Figure 2F). Remarkably, the thermal transition observed in PTETC between 5 and 37 °C was also observed in 14 ACS Paragon Plus Environment

Page 15 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

the monomeric PC fragment although to a lesser degree (Figure 2F) but not in the isolated tetramerization domain PTET (Figure 2E), indicating that the structural changes modulated by temperature were localized within the C-terminal module of P. Modular behavior of P. The possibility to work with complementing fragments and domains provides a simple but effective way to assess whether the modules are structurally independent of each other. Both PN-PTETC and PNTET-PC complementing pairs complete the full length P phosphoprotein. If the spectrum of the mixture of complementing pairs is identical to that of fulllength P, it would indicate that the isolated domains display the same conformation as in the full length protein, supporting that the fragments are structurally independent. As a first control, we found no differences between the CD spectrum of the mixture of both fragment pairs and the arithmetic sum of the individual spectra (Figure 3A-3B, dashed and dotted lines), indicating no interaction between the complementing pairs of fragments. Next, we compared the far-UV CD spectra of the complementing pairs of fragments PN-PTETC (Figure 3A) or PC-PNTET (Figure 3B) with that of full length P at 5 µM. The CD spectrum of the PN-PTETC mixture was identical to that of P (Figure 3A), indicating that PN and PTETC are structurally independent. Conversely, the CD spectrum of the PNTET-PC mixture was substantially different from that of P (Figure 3B, dashed and full lines). We observed a significant increase in the band at 220 nm and a shift in the minimum at 204 nm to 206 nm in full length P compared to the PNTET-PC mixture, which indicated an increase in α-helical content in the context of the full length protein (N= 18 residues) and pointed to the existence of an α-helical structure whose formation requires the presence of the C-terminal module and is stabilized by the tetrameric context. Characterization of a metastable element within the RSV P C-terminal module. Thermal denaturation experiments. To further investigate the conformational stability of the metastable structural elements within P, we carried out thermal denaturation scans monitored by changes in the CD spectral band at 220 nm (Figure 4). Under the experimental conditions, all fragments presented reversible thermal transitions. Thermal denaturation of full length P revealed a structural transition between 5 and 60 °C, with some degree of cooperativity (Figure 4A, upper panel). Within this temperature range PTET, PNTET and PN showed no thermal transition (Figure 4A, lower panel) while PC and PTETC showed transitions similar to that of P (Figure 4B),

15 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 41

in accordance with the results of Figure 2. Between 65°C and 85°C we observed the beginning of a second transition, which corresponds to unfolding of the stable tetramerization domain (Figure 4A). By fitting the thermal scans from 5 to 60 °C to the Gibbs-Helmholtz equation (see “Experimental Procedures”), we estimated the midpoint of the thermal transition (Tm). The Tm estimated for P and PTETC was very similar (25 ± 1 and 26 ± 2 °C respectively). On visual inspection, the PC fragment transition was significantly shifted towards lower temperatures (Figure 4B). This was reflected in a lower value for the estimated Tm of (15 ± 2 °C) although the Tm estimation was less reliable due to the absence of a baseline. These results confirmed that temperature variations within the physiological range can affect the stability of the metastable region mapped within the C-terminal module, and that this metastable structural element was fully contained within the PTETC fragment. On the other hand, the differences in Tm observed between PTETC and PC was consistent with a reduced number of bonds being formed and broken in the transition of the PC fragment, suggesting a lower stability and fewer residues being structured in the metastable element located within PC in the absence of the tetramerization domain. Binding of 8-anilino-1-naphthalenesulfonate (ANS) and chemical unfolding. The ANS fluorescent dye is a sensitive reporter of surface or buried but solvent accessible hydrophobic sites. Since we have previously shown that full-length P binds ANS, and this binding is impaired at 2.0 M Gdm.Cl 28, we hypothesized that ANS binding might report on the metastable structure identified within PTETC. To address this hypothesis, we studied ANS binding of the different fragments relative to full length P. Interestingly, neither PTET, PC nor PN bound ANS in their isolated states, while PTETC bound ANS to the same extent as P (Figure 5A). The fact that ANS bound to tetrameric PTETC but not to PC indicated that the tetrameric context stabilized a tertiary fold, which gave rise to an ANS binding site (Figure 5A). Interestingly, PNTET presented weak ANS binding (Figure 5A), and we hypothesize that this second binding site could map to the αhelical sub-fragment located at the boundary between the N-terminal module and PTET

29, 30.

In

any case, these results clearly demonstrate that the main ANS binding site is located within the P phosphoprotein C-terminal module. We have previously described that the P phosphoprotein presents a three-state Gdm.Cl unfolding transition

28.

The first unfolding transition takes place between 0 and 3.5 M Gdm.Cl where the

16 ACS Paragon Plus Environment

Page 17 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

tetramer remains stable, and the second transition occurs between 4 and 6 M Gdm.Cl (Figure 5B, open circles) and corresponds to the unfolding and dissociation of the tetramerization domain 28. We focused on the 0 to 3.5 M Gdm.Cl transition. PNTET showed no change in secondary structure along the transition, as expected from the absence of consolidated structure within the N-terminal module (Figure 5B, open squares). Conversely, PTETC displayed an unfolding transition that completely overlapped with the transition observed for full length P (Figure 5B, black and open circles), suggesting that the same metastable structural element identified in thermal denaturation experiments was being disrupted. To further address this issue, we analyzed the ability of Gdm.Cl to displace bound ANS as a reporter of the unfolding of the marginally stable structural element and compared it with the secondary structure transition (Figure 5C).

The ANS

displacement in PTETC was identical to that of P, (Figure 5C, black and open squares) indicating that the metastable structure depends on the tetrameric environment of the C-terminal module. Interestingly, ANS was fully displaced before 1.0 M Gdm.Cl (Figure 5C, black and open squares), while at this concentration the change in secondary structure was only at 60% of the full transition (Figure 5C, black and open circles). This observation evidenced the presence of two sub-transitions within the C-terminal module, the first involving loss of hydrophobic and tertiary structure that bound ANS, and the second corresponding to the complete loss of secondary structure of the metastable element. To assess the stabilities and relative magnitudes of the structural change upon chemical unfolding for PTETC and PC, we carried out denaturation experiments along the first transition (0 to 3.5 M) at 10 °C, to stabilize existing structures (Supplementary Figure 3). Both PC and PTETC showed transitions of similar midpoint, indicating similar thermodynamic stability of the structures. However, the number of residues unfolded estimated from the difference in α-helix populations at 4.0 versus 0 M Gdm.Cl was 57 for PTETC and 17 for PC, further suggesting that the C-terminal module contains a structural element with metastable secondary and tertiary components that is formed to a larger extent in the tetrameric context.

Unfolding kinetics of stable and metastable structural elements within RSV P.

17 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 41

We next analyzed the rate at which the C-terminal metastable element was formed or broken. To this end, we performed time trace experiments where we transferred the P phosphoprotein equilibrated at 10 °C to buffer solution equilibrated at 40 °C, and followed secondary structure changes by CD. Upon thermal unfolding, we observed a positive change in ellipticity that occurred within the experimental dead time of mixing,