ISSN : 2278-4586
EISSN : 2278-4608
WAH D.A.1, LAURENTS D.V.2*
1Department of Biochemistry and Molecular Biology, New Jersey Medical School, University of Medicine and Dentistry of New Jersey, 185 South Orange Avenue, Newark, NJ 07101, USA
2Instituto de QuÃmica FÃsica “Rocosolano†Consejo Superior de Investigaciones CientÃficas Serrano 119 28006 Madrid, Spain.
* Corresponding Author : dlaurents@iqfr.csic.es
Received : 15-11-2011 Accepted : 14-02-2012 Published : 17-02-2012
Volume : 1 Issue : 1 Pages : 3 - 8
World Res J Pept Protein 1.1 (2012):3-8
The structural characterization of trypanosome proteins can reveal significant differences relative to their human homologs which may be exploited as novel therapeutic targets. The recent determination of the C-terminal domain of the T. brucei TFIIB transcription factor (tTFIIBc) revealed that its DNA binding surface is much poorer in positive charges compared to the human protein. The T. brucei protein also contains an 18 residue internal segment and a 32 residue C-terminal extension which are not found in human TFIIBc, but their missing electron density barred the determination of their conformations by X-ray crystallography. Here, we have characterized structurally peptides corresponding to these motifs using circular dichroism, fluorescence and nuclear magnetic resonance spectroscopies. The internal segment was found to be mostly disordered, with a slight tendency to adopt helical conformations as evidenced by a modest induction of helicity by 40% TFE and several weak 1HNi-1HNi+1 NOES. The peptide corresponding to the C-terminal extension is essentially disordered. These results show that the lack of electron density was not due to crystallization conditions or packing but to intrinsic disorder. Considering that intrinsically disordered motifs frequently mediate protein-protein interactions, these results are consistent with our proposal that the internal segment and C-terminal extension may compensate for the low positive charge density of T. brucei TFIIB by recruiting auxiliary transcription factors.
Trypanosoma brucei, Sleeping Sickness, NMR Spectroscopy, Intrinsically Disordered Proteins, Circular Dichroism, Fluorescence Spectroscopy.
Sleeping sickness is caused by Trypanosoma brucei. When untreated, it produces irreversible neurological damage and is invariably fatal, claiming some 50000 victims per year. Effective treatments are available but they must be administered by injection and have serious side effects including death in a small percentage of cases [1] . By specifically targeting required proteins, it should be possible to develop more selective therapies with fewer side effects and which could be given orally.
The transcription factor IIB (TFIIB) plays crucial roles in RNA transition initiation. It binds DNA, RNA polymerase II, the TATA box binding protein and organizes other transcription initiation factors [2,3] . Trypanosome TFIIB is a potentially good therapeutic target because it is an essential protein which is distinct from its human counterpart on the level of sequence and tertiary structure; in particular, two striking differences have been recently described for the human and trypanosomal TFIIB protein [4] . First, human TFIIB is much richer in positively charged residues which are strategically positioned to interact with DNA. Second, T. brucei TFIIB has an additional helix, a novel 3/10 helix (H3’A)+loop segment and a 32 residue C-terminal extension. These structures are conserved in Trypanosoma cruzi, the pathogen responsible for Chagas disease, but not in the human protein. The loop segment and the C-terminal extension do not show electron density, either because: 1) the conditions of the crystallization experiment or the packing constraints within the crystal lattice prevent them from adopting a well ordered conformation, or 2) these motifs are intrinsically disordered.
Here, we test between these two possibilities by characterizing structurally peptides corresponding to the helix H3’A + loop and C-terminal segments which are not observed in the X-ray structure using CD, Fluorescence and NMR spectroscopies. We show that both peptides are largely disordered, although the peptide corresponding to the helix H3’A + loop has a slight tendency to adopt helical conformations. Therefore, the absence of electron density for these segments in the tTFIIBc crystal structure is due to a lack of structure and is not a consequence of the crystallization conditions or tTFIIBc packing within the crystal lattice. These results are consistent with our previous proposal that these novel elements may compensate for the lower positive charge density of trypanosomal TFIIB by recruiting additional factors [4] .
Two synthetic peptides were acquired from Genescript Corp. One peptide had the sequence: QTLSW EQGTH ISKGK ESDL corresponding to the short 3/10 helix (residues QTLSW) and the following 14 residues which lack electron density in the X-ray structure, except the Cys was substituted by Ser (underlined). The second peptide had the sequence: DSGEP TAGKR KVDKN SEPEA SGSTK RVKRE ET of the 32 residues at the C-terminus of tTFIIBc which are invisible in the X-ray structure [4] . The Cys present in this second peptide was also substituted by Ser (underlined) to avoid potential problems with cysteine oxidation or disulfide bond formation. These peptides are called the “loop†and “C-terminal†peptides, and correspond to residues 256-274 and 314-345 in the full length tTFIIB protein, respectively. Both peptides were over 95% pure as judged by HPLC and their identities were confirmed by mass spectrometry and NMR spectroscopy.
Putative folded and intrinsically disordered segments in the amino acid residue sequence of the C-domains of TFIIB from T. brucei and T. cruzi were predicted using the program FoldIndex© [http://bioportal.weizmann.ac.il/fldbin/findex] [5] . The Kyte and Doolittle hydrophobicity scale [6] and a residue window size of ten were used.
A Jasco 810 spectropolorimeter was employed for CD spectroscopy. Spectra were recorded over a wavelength range of 190 - 260 nm using 0.5 nm steps, a one nm slitwidth, and a scan speed of 20 nm·min-1. Four scans were acquired and averaged per spectrum. The samples contained 100 mM KCl, 10 mM K2HPO4, 10 mM HAc and 127 mM (loop peptide) or 91 mM (C-terminal peptide). Some spectra also were recorded on samples in 40% TFE, 100 mM KCl, 10 mM KH2PO4 and 10 mM HAc at pH 7. Spectra were obtained at pH 3.5 and 7.0 and at 10.0 and 37.0 ºC. Reference spectra on this buffer alone or buffer with 40% TFE were recorded and subtracted from the spectra recorded on peptide samples.
A Jobin-Yvon Fluoromax 4 instrument was used to obtain fluorescence spectra. The intrinsic fluorescence spectrum of the loop peptide, whose concentration was 61 mM, was measured at 20 ºC in 10 mM K2HPO4, 10 mM HAc buffer, pH 7.0. The excitation wavelength was 280 nm and emission was recorded from 295 to 400 nm. This spectrum and those recorded with ANS were recorded in 1 nm steps scanning at 0.4 s per nm and the slits for excitation and emission were both 2 nm. The excitation and emission wavelengths were calibrated using the Xe lamp excitation maximum and the Raman emission line of water.
ANS, whose concentration was 120 mM, was excited at 365 nm and its emission was recorded over 400 - 650 nm. Spectra were recorded at 20 ºC in buffer at pH 7.0 alone or in the presence of 61 mM loop peptide or 52 mM C-terminal peptide.
All NMR spectra were acquired on a Bruker AV 800 NMR spectrometer equipped with a cryoprobe. Samples contained peptides at 0.3 to 4.5 mM concentration, 100 mM KCl, 5 mM HAc (d3), 5 mM KH2PO4, 1 mM NaN3. Samples also contained 50 mM sodium 2,2 dimethyl-2-silapentane sulfonate as the internal 1H chemical shift reference. This signal was also used to indirectly reference the 13C chemical shifts [7] . Samples were dissolved in 9:1::H2O:D2O for recording 1D 1H and 2D 1H TOCSY (60 ms mixing time), NOESY (150 ms mixing time), and 1H-15N HSQC spectra or 99.9% D2O for recording 2D 1H-13C HSQC spectra. The HSQC spectra were recorded using the natural abundance of 13C and 15N. The pH of the samples was 5.30 to 5.40 except for the initial series of 1D 1H spectra described below. A WATERGATE module was used for solvent suppression for the spectra in 90% H2O and presaturation was employed for samples containing 99.9 % D2O. All spectra were recorded at 10 ºC. Spectra were transformed and processed using the Bruker TOPSPIN program and were assigned with the aid of the program SPARKY [8] .
The amino acid residue composition of the loop segment and C-terminal extension are rich in polar and charged residues and rather poor in larger hydrophobic residues. Such a composition is typical of intrinsically disordered segments (Liu et al. Biochemistry 2006 45:22 6873). Both segments, particularly the C-terminal extension, are predicted to be intrinsically disordered by a sequence-based algorithm FoldIndex© [SupportingFig-1] . Interestingly, similar results are obtained for the TFIIB sequence of T. cruzi.
The secondary structure in the loop and C-terminal peptides was measured by CD spectroscopy. Spectra were measured both in buffer as well as buffered 40% TFE, a co-solvent which is known to enhance helix formation in protein segments possessing an intrinsic propensity to adopt helical conformations [9-11] . The results are shown in [Fig-1] . In aqueous solution, both the loop and C-terminal peptides show features typical of an ensemble of random coil conformations; namely, values close to zero at 222 nm and minima near 200 nm. In the presence of 40% TFE, modest (10 ºC) and slight (37ºC) induction of helical structure is observed in loop peptide’s spectra [Fig-1A] and [Fig-1B] , whereas smaller changes are seen in the C-terminal peptide’s spectra [Fig-1C] and [Fig-1D] .
The fluorescence spectrum of the loop peptide reveals an emission maximum above 350 nm, which is consistent with the Trp indole group being solvent exposed and not buried [Fig-2A] . The loop and C-terminal peptides do not enhance the fluorescence of ANS [Fig-2B] . This suggests that they have no partly ordered, solvent exposed hydrophobic patches [12] .
The peptides were dissolved in one volume of milliQ H2O, then one volume of 2x buffer. The peptides dissolved cleanly with no sign of turbidity. An initial series of 1D 1H spectra were recorded for each peptide at low concentration (0.3 to 0.5 mM) at pH 3, 5, 7 and 10. Sharp lines and low dispersion characteristic of unfolded peptides were observed [SupportingFig-2] . At alkaline pH, the 1HN signals disappear within 30 minutes due to rapid exchange of these protons with solvent (data not shown). This suggests that these peptides lack stable intramolecular hydrogen bonds and therefore that they lack stable helices and beta sheets.
The 2D 1H TOSCY and NOESY spectra of the loop and C-terminal peptides were assigned by standard strategies [13] . These assignments were corroborated and, in the case of some of the longer side chains, completed by analysis of the 1H-13C HSQC and 1H-15N HSQC (C-terminal peptide only) spectra [Fig-3] . The assignments for the loop and C-terminal peptides are listed in [Table-1] and [Table-2] , respectively. No significantly upfield shifted peaks, which generally arise from the packing of aliphatic groups on aromatic rings, were observed. All the 1HN signals occupy a narrow range of chemical shift values at 10 ºC (7.9 - 9.0 ppm). These observations are hallmarks of peptides lacking preferred conformations.
Spectra were recorded on 1 - 4 mM peptides samples in 100 mM KCl, 5 mM K2HPO4, 5 mM HAc, 1 mM NaN3 buffer, pH 5.3 at 10 ºC in 90% H2O, 10% D2O, except B, which was recorded in 99.9 % D2O buffer.
The NOESY spectrum recorded on the loop peptide shows some signs of minor populations of ordered conformations; namely, there are two weak crosspeaks arising from the Trp and a Leu or Ile side chain (data not shown). These signals likely arise from medium range contacts between W5 and L3. There are several HNi - HNi+1 NOE crosspeaks arising from residues G8, T9, H10 and I11; and E16, S17 and D18 of the loop peptide [SupportingFig-3] . Such signals are characteristic of turn or helical structure. This turn or helical structure is not significantly stable since as previously noted, the NH readily exchange at higher pH values. Additional HNi - HNi+1 NOE crosspeaks from W5, E6 and Q7; and S12, K13 and G14 might be present but can not be observed due to their proximity to the diagonal. No such signs of even minor populations of ordered structure are seen in the C-terminal peptide.
The population of partial secondary structure was assessed on the level of individual residues by comparing the experimental 1Ha and 13Ca values with reference “random coil†values [14] . Conformation chemical shift (Dd = d exp – d r.c.) values of -0.40 ppm for 1Ha and +3.10 ppm for 13Ca are expected for 100% helical structures, and values of -0.10 ppm for 1Ha and +0.70 ppm for 13Ca are commonly used as thresholds for identifying partial structure formation. Similar Dd values but with opposite signs are used for detecting and quantifying b-structures. Except for the terminal residues, whose chemical shift values are altered by end charge effects, the Dd values are small and below the thresholds for detecting structure in the C-terminal peptide [Fig-4] . For the loop peptide, a weak tendency towards helical conformations was detected for residues E6 and Q7.
Taking into account all the data, we conclude that the C-terminal peptide does not have a preferred conformation. Therefore, the absence of electron density for the C-terminal segment in tTFIIBc is due to a lack of structure and is not a consequence of the conditions of the crystallization experiments or packing of tTFIIBc within the crystal lattice in such a way that impedes the correct folding of these segments. The loop peptide is also largely disordered but has a slight tendency to adopt helical conformations. Interestingly, the N-terminal region of this peptide (residues Q1-T2-L3-S4-W5; corresponding to Q256-T257-L258-S259-W260, in the full-length protein) adopts a short 3/10 helix in the crystal structure of tTFIIBc. This strongly suggests that interactions between this segment, such as hydrophobic contacts between L258 and L218 of helix H2’ and among W260 and Q309 and F312 of helix H6’, drive its folding.
Intrinsically disordered motifs, domains and proteins are useful in macromolecular interactions when speed, a relaxed specificity and binding to multiple partners are needed [15] . Intrinsically disordered domains appear to be relatively common in proteins that regulate transcription [16] For example, the transcription of the toxin-antitoxin pair DOC/PHD in bacteria is regulated by the allosteric binding of an intrinsically unfolded domain [17] and moreover, the interferon-regulated transcription factor IRF-1, which is also a tumor suppressor, contains a disordered domain that serves as a binding interface for several regulatory proteins [18] .
The intrinsic disorder of the tTFIIB internal loop and C-terminal extension reported here supports our proposal that they could serve as binding sites for additional factors that aid transcription initiation in Trypanosoma brucei [4] . These segments could be carrying out similar functions in T. cruzi since they are predicted by sequence analysis to also be intrinsically disordered. T. cruzi is the causative agent of Chagas disease, a debilitating and sometimes fatal illness that affects 16 - 18 million Latin Americans, for which no vaccine and no completely effective treatment are available. As these polypeptide segments are absent in human TFIIB, they can be considered to be useful targets for developing more selective and effective therapeutics.
This work was supported by a grant from the Spanish Ministry of Science and Innovation Grant CTQ2010-21567-C02-02 (to D.V.L). D.A.W is an employee of the Howard Hughes Medical Institute. We thank Prof. Marta Bruix for critically reading the manuscript and Dr. David Pantoja for NMR technical assistance.
[1] WHO (2010) Sleeping Sickness.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[2] Li Y., Flanagan P.M., Tschochner H., Kornberg R.D. (1994) Science, 263,805-807.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[3] Bushnell D.A., Westover K.D., Davis R.E., Kornberg R.D. (2004) Science, 303, 983-988.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[4] Ibrahim B.S., Kanneganti N., Rieckhof G.E., Das A., Laurents D.V., Palenchar J.B., Bellofatto V., Wah D.A. (2009) Proc. Natl, Acad. Sci. USA.
, 106(32),13242-13247.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[5] Prilusky J., Felder C.E., Zeev-Ben-Mordehai T., Rydberg E.H., Man O., Beckmann J.S., Silman I., Sussman J.L. (2005) Bioinformatics, 21(16),3435-3438.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[6] Kyte J., Doolittle R.F. (1982) J. Mol. Biol., 157(1),105-132.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[7] Markley J.L., Bax A., Arata Y., Hilbers C.W., Kaptein R., Sykes B.D., Wright P.E., Wüthrich K. (1998) Pure Appl. Chem., 70(1),117-142.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[8] Goddard T.D. and Kneller D.G., SPARKY.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[9] Rico M., Nieto J.L., Santoro J., Bermejo F.J., Herranz J., Gallego E. (1983) FEBS Lett.,162, 314-319.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[10] Dyson H.J., Merutka G., Waltho J.P., Lerner R.A., Wright P.E. (1992) J. Mol. Biol., 226, 795-817.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[11] Dyson H.J., Sayre J.R., Merutka G., Shin H.C., Lerner R.A. Wright P.E. (1992) J. Mol. Biol., 226, 819-835.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[12] Semisotnov G.V., Rodionova N.A., Razgulyaev O.I., Uversky V.N., Gripas A.F., Gilmanshin R.I. (1991) Biopolymers, 31,119-128.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[13] Wüthrich K. (1986) NMR of Proteins and Nucleic Acids.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[14] Wishart D.S., Sykes B.D. (1994) J. Biomol. NMR, 4,171-180.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[15] Pontius B.W. (1993) Trends Biochem. Sci., 18(5), 181-186.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[16] Liu J., Perumal N.B., Oldfield C.J., Su E.W., Uversky V.N. Dunker A.K. (2006) Biochemistry, 45(22), 6873-6888.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[17] Garcia-Pino A., et al. (2010) Cell, 142(1), 101-111.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
[18] Narayan V., Halada P., Hernychová L., Chong Y.P., Žáková J., Hupp T.R., Vojtesek B., Ball K.L. (2011) J. Biol Chem., 286(16), 14291-14303.
» CrossRef » Google Scholar » PubMed » DOAJ » CAS » Scopus
 | Fig. 1- CD Spectra of tTFIIB Peptides A. Loop peptide, 10 ºC. open circles and solid line, pH 3.5; open squares and dotted line, pH 7.1; filled diamonds and solid line, pH 7.1 with 40% TFE. B. Loop peptide, 37 ºC. the symbols have the same meaning as panel A. C. C-terminal peptide, 10 ºC. open triangles and solid line, pH 3.5; inverted open triangles and dotted line, pH 6.8; filled triangles and solid line, pH 6.8 with 40% TFE. D. C-terminal peptide, 37 ºC. the symbols have the same meaning as panel C. |
 | Fig. 2- Intrinsic and ANS Fluorescence Spectra of tTFIIB peptides A. Fluorescence spectrum of the loop peptide recorded at pH 7.0, 20 ºC. B. Fluorescence spectra of ANS alone in buffer or in the presence of the tTFIIB peptides recorded at pH 7.0, 20ºC. |
 | Fig. 3- Assignment of the NMR spectra of tTFIIBc peptides A. Regions of the 2D 1H TOCSY (black) and NOESY (red) spectra of the loop peptide. The 1HN-1Ha crosspeaks are labeled. The broken red line shows the sequential connectivities of residues K15 to L19 via NOE signals. B. 13Ca-1Ha region of the HSQC spectrum of the loop peptide. The Ser and Thr 13Cb-1Hb resonances are also labeled. C. The 15N-1H HSQC spectrum of the C-terminal peptide, showing the assigned signals. The Arg side chain peak at 7.27 ppm (1H) is folded in the 15N dimension. |
 | Fig. 4- Backbone Conformational Shifts for the loop (A) and C-terminal peptides (B) at 10 ºC in pH 5.3 buffer with 100 mM KCl. Dd values for 1Ha (gray bars) and 13Ca (black bars) are shown. |
 | Supporting Fig.1-Bioinformatics Analysis of the T. brucei (top panel) and T. cruxi TFIIB C-terminal domain sequences using FoldIndex©. |
 | Supporting Fig. 2- 1D 1H NMR Spectrum of the tTFIIBc loop peptide in 100 mM KCl, 5 mM K2HPO4, 5mM HAc, 3 mM NaN3, pH 5.3, 90% H2O, 10% D2O at 10ºC. |
 | Supporting Fig. 3- 2D 1H NOESY NMR Spectrum of the tTFIIBc loop peptide in 100 mM KCl, 5 mM K2HPO4, 5mM HAc, 3 mM NaN3, pH 5.30% H2O, 10% D2O at 10ºC. 1HNi-1HNi+1 NOE crosspeaks indicative of tight turn or helical structure are labeled. |
 | Table 1- NMR Chemical Shifts of the cTFIIB loop domain peptide 1H chemical shifts are shown in plain text and 13C chemical shifts are in bold text. |
 | Table 2- NMR Chemical Shifts of the cTFIIB C-terminal domain 1H chemical shifts are shown in plain text, 13C chemical shifts are in bold text, and 15N chemical shifts are shown in italics. |