1 3

Eur Biophys J DOI 10.1007/s00249-014-0967-8

OrIgInal PaPEr

Chagas disease: a homology model for the three‑dimensional structure of the Trypanosoma cruzi ribosomal P0 antigenic protein

Juan Arturo Gomez Barroso · Carlos Fernando Aguilar

received: 30 December 2013 / revised: 16 May 2014 / accepted: 20 May 2014 © European Biophysical Societies’ association 2014

Keywords Trypanosome cruzi · P0 ribosomal protein · Three-dimensional structure · Chagas disease

Introduction

ribosomal P proteins are a complex of proteins that form a long and protruding region, called the stalk, in the large subunit of the ribosome. In eukaryotes, the P fam-ily encompasses protein P0 (34kD), P1 and P2 (~10 kD) (liljas 1991). a third protein, P3, has been found in plants (Bailey-Serres et al. 1997). The number of proteins in the P1 and P2 families varies among species. In mammals, the complex is formed by P0 and two copies of P1 and P2, but Saccharomyces cerevisiae and T. cruzi, the parasite that causes Chagas disease, have five members: P0, P1α, P1β, P2α and P2β. all of them have a conserved acidic motif in the C-terminal end and show high sequence identity with the P0 C-terminal (last 100 amino acids). In prokaryotes (bacteria Thermotoga maritima), the ortholog complex is formed by the protein l10 and multiple copies of l7/l12 bound in the region of Domain II of 23S rrna. It inter-acts with 28S rrna at the thiostrepton loop, protein l11 and laterally with the gTPase domain of the elongation factor EF-g (Diaconu et al. 2005). There is no significant sequence identity between the eukaryote and bacterial com-plex, although it is generally accepted that both function in a similar way. They participate directly in the translocation step of protein synthesis through the recruitment of trans-lation factors to the ribosome and the stimulation of gTP hydrolysis of the ribosome-bound factors by way of stabi-lization of their active conformation (Diaconu et al. 2005). P0 is the minimal portion of the stalk, which is able to sup-port accurate protein synthesis, but at a lower rate than the complete stalk (Santos and Ballesta 1995).

Abstract ribosomal P proteins form a “stalk” complex in the large subunit of the ribosomes. In Trypanosoma cruzi, the etiological agent of Chagas disease, the complex is formed by five P protein members: TcP0, TcP1α, TcP1β, TcP2α and TcP2β. The TcP0 protein has 34 kDa, and TcP1 and TcP2 proteins have 10 kDa. The structure of T. cruzi P0 and the stalk complex TcP0–TcP1α–TcP1β–TcP2α–TcP2β have not been solved to date. In this work, we constructed a three-dimensional molecular model for TcP0 using homol-ogy modeling as implemented in the MODEllEr 9v12 software. The model was constructed using different tem-plates: the X-ray structures of the protein P0 from Piro-coccus horikoshii, a segment from the Danio renio Ca+2/K+ channel and the C-terminal peptide (C13) from T. cruzi ribosomal P2 protein; the Cryo-EM structure of Triticum aestivum P0 protein and the nMr structure of Homo sapi-ens P1 ribosomal protein. TcP0 has a 200-residue-long n-terminal, which is an α/β globular stable domain, and a flexible C-terminal, 120-residue-long domain. The molecu-lar surface electrostatic potential and hydrophobic surface were calculated. The surface properties are important for the C-terminal’s antigenic properties. They are also respon-sible for P0-specific binding to rna26S and the binding to the P1–P2 proteins. We explored and identified protein interactions that may be involved in conformational stabil-ity. The structure proposed in this work represents a first structural report for the TcP0 protein.

J. a. gomez Barroso (*) · C. F. aguilar laboratorio de Biología Molecular Estructural; Facultad de Química, Bioquímica y Farmacia, Universidad nacional de San luis, San luis, argentinae-mail: jagomez@unsl.edu.ar

C. F. aguilar e-mail: carlosfaguilar@gmail.com

Eur Biophys J

1 3

Previous studies including circular dichroism, tryptophan fluorescence and limited proteolysis done in our laboratory suggested that P0 has a compact, stable, trypsin-resistant, 200-residue-long n-terminal domain belonging to the α/β class and a more flexible, degradable, helical, ~120-residue-long C-terminal domain (Juri ayub et al. 2005). These facts were confirmed in part for the crystal structures of the n-ter-minal domains of the archaea l10E (H. marismortui) and bacterial l10 (Thermothoga maritima) (Diaconu et al. 2005), which showed striking cross-kingdom structural similarities, although they did not have a significant sequence identity. The TcP0 C-terminal (~120 residues long) is the TcP1/TcP2-binding domain. It has a pairwise sequence identity with TcP1 and TcP2 proteins ranging from 40 % (TcP0–TcP2β) to 33 % (TcP0-TcP1β). If local alignment is performed, the identity goes up to 60 %. In contrast, bacterial l10 is much shorter than TcP0 and does not have a common domain with l7/l12. The C-terminal domains of P0 and the P1–P2 proteins are thought to belong to a new class called natively unfolded or intrinsically disordered proteins (IDPs). The classical protein structure-function paradigm, which can be defined in a few words as the amino acid sequence of a protein determining a stable and defined 3D structure, which is then responsible for the function, has been extended and improved by evidence showing that there is an alternative native state in which proteins exist and function despite lacking a well-defined folded three-dimensional structure (Tompa 2002, 2012; Uver-sky 2002). IDPs are abundant in living organisms, and they accomplish essential functions such as basic regulatory roles in cellular processes. They have been classified in six broad categories according to their functional modes. ribosomal bacterial proteins l7/l12, orthologs of the P1 and P2 family, belong to the assembler category (Tompa 2005).

It has been suggested that these proteins act as a molecu-lar switch in the ribosome, performing ratchet-like motions related to the binding and release of elongation factors and gTP hydrolysis. This hypothesis was based on the differ-ences between the nMr structure of l7/l12 in solution and the crystal structure in which the hinge region of the protein goes from a flexible, unstructured coil, found in solution, to an α-helix that associates with the n-terminal domain and brings together the C-terminal domain to form a compact protein (Bocharov et al. 2004). a great deal of experimental evidence suggests that T. cruzi P1–P2 pro-teins and the C-terminal domain of P0 would behave in the same manner. They have extreme proteolytic sensitivity and greater mobility than expected in SDS-PagE gel elec-trophoresis (Juri ayub et al. 2005).

Far-UV circular dichroism (CD) spectroscopy studies of yeast P1α and P2β under different conditions (pH, tem-perature, urea concentration) show that the isolated pro-teins have about 60 % helical secondary structure and lack a well-defined tertiary structure. The P1α–P2β complex has

a well-packed hydrophobic core indicating that the protein complex has gained a stable tertiary structure (Tchórzewski et al. 2003). Using urea transversal gradient gels, our labo-ratory has demonstrated that T. cruzi P1α and P2β behave in the same way as S. cerevisiae orthologs. P0, due to its α/β domain, behaves in a different manner in this test, showing a sigmoid denaturing pattern typical of a globular protein (Juri ayub et al. 2001). Interaction studies using the yeast two-hybrid system reveal that the P0 C-terminal is able to interact with all P1–P2 proteins. a quantitative evaluation of the interactions indicates that the strongest interaction is with TcP1α (Juri ayub et al. 2005a, b). TcP0 has two bind-ing sites for P1/P2. TcP1 and TcP2 proteins form P1α–P2β/P1β–P2α pairs. analysis of the interaction of the truncated variants of the P0 C-terminal domain shows that the region responsible for binding the P proteins is a relatively short stretch of 19 amino acids (254–272), which are positioned before an alanine-rich region (Juri ayub et al. 2005a, b). The P0 C-terminal peptide is highly antigenic and a major target of the antibody response in patients suffering from chronic heart disease produced by the T. cruzi parasite (lopez Ber-gami et al. 2001). Our laboratory is working on the X-ray structure of T. cruzi P0 and the stalk complex P0–P1α–P1β–P2α–P2β, but we so far have been unable to solve them. The aim of this work is to obtain a homology model for TcP0 to characterize the structural and functional domains.

Homology modeling

The sequence for T. cruzi 60S acidic ribosomal protein P0 was obtained from the nCBI database (national Center for Biotechnology Information). P0 has 323 amino acid residues, and the accession number is EKg03452. Comparative mod-eling was done by searching the Protein Data Bank (PDB) for known protein structures using the target sequences with the sequence of P0 as a query. This step was realized using BlaSTp, and the results provide potential templates for dif-ferent primary structural regions. The templates were aligned with the TcP0 sequence using the ClUSTal W (Jeanmougin et al. 1998) program. The full protein was modeled using five PDB-related and -unrelated protein templates. They have the Cryo-EM 5.5 Ǻ structure of Triticum aestivum P0 protein, 3IZr_s (armache et al. 2010); rMn structure of Homo sapi-ens P1 ribosomal protein, 4BEH_a (lee et al. 2013); X-ray 1.89 Ǻ crystallographic structure of Trypanosome cruzi C13 ribosomal P2 peptide, 3SgE_K (Pizarro et al. 2011); X-ray 3.61 Ǻ crystallographic structure of Danio renio Ca+2/K+ channel, 3U6n_a (Yuan et al. 2011) and X-ray 2.13 Ǻ crys-tallographic structure of Pirococcus horikoshii P0 protein, 3a1Y_g (naganuma et al. 2010) (Fig. 1). These alignments were used for the TcP0 homology modeling using MODEl-lEr 9v12 (Sali and Blundell 1993).

Eur Biophys J

1 3

Molecular dynamics

Molecular dynamics studies were realized using the serv-ers for I-Tasser (roy et al. 2010; Zhang 2008) and evfold (Marks et al. 2012) to observe the behavior of the struc-tured and unstructured regions found in the TcP0 homol-ogy model. I-Tasser was first run using the full sequence for TcP0. a second run was done with the C-terminal trun-cated P0 (a261-F323 region).

Analysis of the model

The charge distribution was calculated using aPBS as implemented in Pymol (Delano 2002). Chimera (Pettersen et al. 2004) was used to calculate the hydrophobic surface. These surfaces would allow identifying potential electro-static and/or hydrophobic patches that would bind different molecules or macromolecules. The intra-protein interac-tions were identified with the protein interaction calculator (PIC) server (Tina et al. 2002) and the COnTaCT program (CCP4). The quality of the models was assessed using the Coot (Emsley et al. 2012) and Procheck (laskowski et al. 1993) programs. Visualization and structural analyses of the final protein model were carried out with the Pymol and Chimera programs.

Results and discussion

Homology modeling of TcP0

The TcP0 model can be concisely described as having a structured n-terminal domain associated with a flex-ible, unstructured C-terminal domain. The n-terminal is

composed of a globular αβ domain (214 residues) joined to an α domain formed by two α-helices (48 residues). The globular αβ domain belonged to the α–β complex architecture according to CaTH (Fig. 2). The n-terminal αβ domain can be described as having two subdomain regions linked by two small anti-parallel β-chains. This type of link has been classified as a small two-linker β-sheet (george and Heringa 2002). The first subdo-main region is formed by five anti-parallel β-chains (β1, β2, β3, β12 and β13) and seven α-helices (α1, α2, α3, α4, α5, α6 and α9) ordered on both sides of the sheet. The α5 α-helix and β3 β-chain are joined by a 26-res-idue-long loop. The second subdomain is composed of three small β-sheets (β5–β10, β6–β9 and β7–β8) and two

Fig. 1 The schematic repre-sentation shows the different template sequences used (green ribbons) on Modeller assays that align with the TcP0 primary structure (orange ribbon). The table shows the identity and positive percentage values observed on alignments realized in the Modeller assay

Fig. 2 Structure of TcP0 model: The figure shows three domains: the n-terminal αβ domain (blue), the α domain (green) and the C-termi-nal domain (red). The yellow arrow indicates the apparent proteolytic site

Eur Biophys J

1 3

α-helices (α7 and α8). Each of the sheets is formed by two antiparallel β-chains. The α domain is assembled by α-helices α10 and α11 separated by a six-residue-long loop (Fig. 3). This domain is responsible for binding to the P1 and P2 proteins (Smulski et al. 2011). Finally, the C-terminal flexible domain has no secondary structural elements.

Electrostatic and hydrophobic surface

The n-terminal domain showed two main positive patches on its surface. The first is formed by helices α1 and α4 and the second by helices α7 and α8. Instead, the C-terminal domain displayed zones with negative density charges. The main negatively charged zone is formed by the last 13 residues and

Fig. 3 Shows the topology dia-gram for TcP0. The n-terminal domain is colored blue and green (the two subdomains of the αβ domain), and the α domain is colored yellow. The unstructured C-terminal domain is represented by the red line

Fig. 4 Electrostatic surface of TcP0 (colored red to blue rang-ing from the most negative to most positive charged regions; neutral surfaces are shown in white). a The electrostatic sur-face of c (the ribbon represents TcP0). The green arrows point to the main charged zones (N1, 2, 3 and P1, 2). b and d are the a and c pictures rotated 90°

Eur Biophys J

1 3

is the only structured region in the C-terminal domain. These residues form a U-shaped tail. The helices in the α domain showed positively and negatively charged regions along their surfaces. The main hydrophobic zone, placed on the α domain, is the binding region to the P1 and P2 proteins. The other zone is located in the αβ domain (Fig. 4).

Validation

The ramachandran plot showed that 87.1 % of the phi and psi angles are in the most favored conformations, 10.8 % in the allowed conformation, 1 % in the generously allowed conformation and only three amino acids in the disallowed region (Fig. 5).

Molecular dynamics

The five structural models of the TcP0 protein obtained using the I-Tasser server showed a n-terminal globular α–β

domain. The five models obtained for the C-terminal trun-cated TcP0t showed more flexibility and a lower C-score value than the full protein (Fig. 6). The 11 models obtained with the evfold program showed a 140-residue-long n-ter-minal globular domain and a C-terminal flexible domain.

Conclusions

The constructed 3D model of TcP0 showed a 214-resi-due globular domain with an α–β complex architecture as defined by CaTH. reassuringly, the size of this domain is compatible with the result of previous limited proteolysis assays made in our laboratory, which indicated that the first 200 amino acids would form a structural domain with high contents of beta strands and alpha helices (Juri ayub et al. 2005). also, the proteolysis site found experimentally in the region of D215, D216, E219 and r220 coincides with the linker region between the α–β and alpha domain (Fig. 1).

Furthermore, the model shows two regions of positive charge density that are in the region located favorably to be the binding sites for the rna26S. The alpha domain formed by the two alpha helices is also in a favorable posi-tion to bind TcP1α, TcP1β, TcP2α and TcP2β.

Finally, the C-terminal unstructured and flexible domain formed by 61 amino acids possesses a highly negative charge density region at the end, which is compatible with the hypothesis proposing that the function of this highly flexible and charged domain is to fetch elongation factors. This observation was confirmed by the I-Tasser and evfold results. The C-terminal negative charge end is responsi-ble for the antigenic protein properties, which produce the autoantibodies responsible for Chagas disease.

The three-dimensional model proposed in this work pro-vides the first TcP0 structure described to date. This infor-mation is very important for understanding the structure-function of the T. cruzi ribosome and the ribosomal stalk. Furthermore, the study of this protein is of fundamental importance because of its involvement in the pathogenesis of Chagas disease.Fig. 5 ramachandran plot for TcP0 calculated by PrOCHECK

Fig. 6 I-Tasser C score, TM score and rMSD values for five models corresponding to the TcP0 full and truncated forms. The lowest values for the truncated form indicate higher disorder than the full TcP0

Eur Biophys J

1 3

Acknowledgments This work was supported by the Secretaría de Ciencia y Técnica de la Universidad nacional de San luis, argentina.

References

armache JP, Jarasch a, anger aM, Villa E, Becker T, Bhushan S, Jossinet F, Habeck M, Dindar g, Franckenberg S, Marquez V, Mielke T, Thomm M, Berninghausen O, Beatrix B, Söding J, Westhof E, Wilson Dn, Beckmann r (2010) localization of eukaryote-specific ribosomal proteins in a 5.5-Å cryo-EM map of the 80S eukaryotic ribosome. Proc natl acad Sci USa 107(46):19754–19759. doi:10.1073/pnas.1009999107

Bailey-Serres J, Vangala S, Szick K, lee CH (1997) acidic phospho-protein complex of the 60 S ribosomal subunit of maize seedling roots. Components and changes in response to flooding. Plant Physiol 114:1293–1305

Bocharov EV, Sobol ag, Pavlov KV, Korzhnev DM, Jaravine Va, gudkov aT, arseniev aS (2004) From structure and dynamics of protein l7/l12 to molecular switching in ribosome. J Biol Chem 279:17697–17706

Delano Wl (2002) The PyMOl molecular graphics system. Delano Scientific, San Carlos. http://www.pymol.org

Diaconu M, Kothe U, Schlunzen F, Fisher n, Harms JM, Tonevitsky ag, Stark H, rodnina MV, Wahl MC (2005) Structural basis for the function of the ribosomal l7/l12 stalk in factor binding and gTPase activation. Cell 121:991–1004

Emsley P, lohkamp B, Scott Wg, Cowtan K (2012) Features and development of Coot. acta Crystallogr D Biol Crystallogr 66(Pt 4):486–501. doi:10.1107/S0907444910007493

george ra, Heringa J (2002) an analysis of protein domain link-ers: their classification and role in protein folding. Protein Eng 15(11):871–879

Jeanmougin F, Thompson JD, gouy M, Higgins Dg, gibson TJ (1998) Multiple sequence alignment with clustal X. Trends Bio-chem Sci 23:403–405

Juri ayub M, levin MJ, aguilar CF (2001) Overexpression and refolding of the hydrophobic ribosomal P0 protein from Trypano-soma cruzi: a component of the P1/P2/P0 complex. Protein Expr Purif 22:225–233

Juri ayub M, gómez Barroso a, levin MJ, aguilar CF (2005a) Pre-liminary structural studies of the hydrophobic ribosomal P0 Pro-tein from Trypanosoma cruzi: a part of the P1/P2/P0 complex. Protein Pept lett 12:521–526

Juri ayub M, Smulski Cr, nyambega B, Bercovich n, Masiga D, Vazquez MP, aguilar CF, levin MJ (2005b) Protein–protein interaction map of the Trypanosoma cruzi ribosomal P protein complex. gene 357(2):129–136

laskowski ra, Macarthur MW, Moss DS, Thornton JM (1993) PrOCHECK—a program to check the stereochemical quality of protein structures. J appl Cryst 26:283–291

lee KM, Yusa K, Chu lO, Yu CW, Oono M, Miyoshi T, Ito K, Shaw PC, Wong KB, Uchiumi T (2013) Solution structure of human P1*P2 heterodimer provides insights into the role of eukaryotic stalk in recruiting the ribosome-inactivating protein trichosanthin

to the ribosome. nucleic acids res 18:8776–8787. doi:10.1093/nar/gkt636

liljas a (1991) Comparative biochemistry and biophysics of riboso-mal proteins. Int rev Cytol 124:103–136

lopez Bergami P, Scaglione J, levin MJ (2001) antibodies against the carboxyl-terminal end of the Trypanosoma cruzi ribosomal P proteins are pathogenic. FaSEB J 15:2602–2612

Marks DS, Hopf Ta, Sander C (2012) Protein structure predic-tion from sequence variation. nat Biotechnol 30:1072–1080. doi:10.1038/nbt.2419

naganuma T, nomura n, Yao M, Mochizuki M, Uchiumi T, Tanaka I (2010) Structural basis for translation factor recruitment to the eukaryotic/archaeal ribosomes. J Biol Chem 285(7):4747–4756. doi:10.1074/jbc.M109.068098

Pettersen EF, goddard TD, Huang CC, Couch gS, greenblatt DM, Meng EC, Ferrin TE (2004) UCSF Chimera—a visualization system for exploratory research and analysis. J Comput Chem 25(13):1605–1612

Pizarro JC, Boulot g, Bentley ga, gómez Ka, Hoebeke J, Honte-beyrie M, levin MJ, Smulski Cr (2011) Crystal structure of the complex mab 17.2 and the C-terminal region of Trypanosoma cruzi P2ß protein: implications in cross-reactivity. PloS negl Trop Dis 5(11):e1375. doi:10.1371/journal.pntd.0001375

roy a, Kucukural a, Zhang Y (2010) I-TaSSEr: a unified platform for automated protein structure and function prediction. nat Pro-toc 4:725–738

Sali a, Blundell Tl (1993) Comparative protein modelling by satis-faction of spatial restraints. J Mol Biol 234(3):779–815

Santos C, Ballesta JP (1995) The highly conserved protein P0 car-boxyl end is essential for ribosome activity only in the absence of proteins P1 and P2. J Biol Chem 270:20608–20614

Smulski Cr, longhi Sa, ayub MJ, Edreira MM, Simonetti l, gómez Ka, Basile Jn, Chaloin O, Hoebeke J, levin MJ (2011) Interac-tion map of the Trypanosoma cruzi ribosomal P protein complex (stalk) and the elongation factor 2. J Mol recognit 2:359–370

Tchórzewski M, Krokowski D, Boguszewska a, liljas a, grankowski n (2003) Structural characterization of yeast acidic ribosomal P proteins forming the P1α–P2β heterocomplex. Biochemistry 42:3399–3408

Tina Kg, Bhadra r, Srinivasan n (2002) PIC: protein interactions calculator. nucleic acids res 35:W473–W476

Tompa P (2002) Intrinsically unstructured proteins. Trends Biochem Sci 27(10):527–533

Tompa P (2005) The interplay between structure and function in intrinsically unstructured proteins. FEBS lett 579:3346–3354

Tompa P (2012) Intrinsically disordered proteins: a 10-year recap. Trends Biochem Sci 37(12):509–516. doi:10.1016/j.tibs.2012.08.004

Uversky Vn (2002) What does it means to be natively unfolded? Eur J Biochem 269:2–12

Yuan P, leonetti MD, Hsiung Y, MacKinnon r (2011) Open structure of the Ca2+ gating ring in the high-conductance Ca2+-activated K+ channel. nature 481(7379):94–97. doi:10.1038/nature10670

Zhang Y (2008) I-TaSSEr server for protein 3D structure prediction. BMC Bioinform 9:40