Amyloid Fiber Formation by Synthetic Peptides ...

2 downloads 88 Views 341KB Size Report
The major curlin subunit, CsgA, spontaneously forms amyloid fibers [5]. CsgA harbours five successive repeating units of about twenty amino acids R1 (43- ...
Send Orders of Reprints at [email protected] Protein & Peptide Letters, 2013, 20, 0000-0000

1

Amyloid Fiber Formation by Synthetic Peptides Derived from the Sequence of the Protein CsgA of Escherichia coli Pierre Lembré, Charlotte Vendrely and Patrick Di Martino* Laboratoire ERRMECe-EA1391, Institut des matériaux-FD4122, Université de Cergy-Pontoise, 2, av Adolphe Chauvin, 95302 Cergy-Pontoise Cedex, France Abstract: We characterized the formation of amyloid fibers by two peptides derived from the CsgA sequence: R5 (133151) corresponding to the whole repeating unit R5 and a truncated form of this peptide called R5T (134-143). In the presence of either of the two peptides: an increase in the fluorescence intensity of Thioflavin T was observed; a shift of the absorbance of Congo red was measured; spontaneous formation of amyloid fibers was observed by polarized light as well asatomic force microscopy imaging. Large-size aggregates were observed with R5 while R5T formed fagots of individualized fibers. The infrared spectroscopy analysis revealed the presence of a greater number of intermolecular bonds for R5.In conclusion, a10 aminoacids peptide derived from the R5 sequence was sufficient for the spontaneous formation of amyloid fibrils but not to form large-size aggregates of fibers.

Keywords: Amyloïd, biofilm, CsgA, Curli, fiber, peptide. INTRODUCTION Amyloid fibers are ordered protein aggregates rich in sheets and assembled through intermolecular hydrogen bonds [1]. The formation of amyloid fibers to the bacterial surface has been observed in a large number of phyla in Proteobacteria, Bacteriodetes, Chloroflexi, and Actino bacteria [2]. In Escherichia coli, filamentous structures called curli form amyloid fibers that have a role in bacterial adhesion to biological and abiotic surfaces and in biofilm formation [3,4]. The major curlin subunit, CsgA, spontaneously forms amyloid fibers [5]. CsgA harbours five successive repeating units of about twenty amino acids R1 (43-SELNIYQYG GGNSALALQTDARN-65), R2 (66-SDLTITQHGGGNGA DVGQGSDD-87), R3 (88-SSIDLTQRGFGNSATLDQWN GKN-110), R4 (111-SEMTVKQFGGGNGAAVDQTASN132), and R5 (133-SSVNVTQVGFGNNATTAHQY-151) that contain the consensus sequence SX5QXGX2NXAX3Q [5,6]. The same consensus sequence is present in the repeated units of AgfA, the CsgA homolog from Salmonella [6]. The R1 and R5 units contain sequences that contribute significantly to the ability of CsgA to bind human proteins such as fibronectin, plasminogen, tissue plasminogen activator, and 2-microglobulin [7]. Synthetic peptides overlapping R1 and R5 units have been shown to bind in vitro to these human proteins. Each repeating unit is predicted to form a strand-loop-strand motif that closely resembles the crossspine structure described for many disease-associated amyloids. Peptides corresponding to any of these five repeating units (R1, R2, R3, R4, and R5) can form fibers in vitro [5,8]. Nevertheless, amyloido genesis requires a 10 times higher *Address correspondence to this author at the Laboratoire ERRMECeEA1391, Institut des matériaux-FD4122, Université de Cergy-Pontoise, 2, av Adolphe Chauvin, 95302 Cergy-Pontoise Cedex, France; Tel: +33134256606; Fax: +33134256694; E- mail: [email protected] 0929-8665/13 $58.00+.00

concentration of peptides for R2 or R4 than for R1, R3 or R5. The aim of this study was to determine if the entire sequence of the R5 repeating unit containing the consensus sequence SX5QXGX2NXAX3Q is needed for the formation of amyloid fibers in vitro. A combination of technical approaches was used to look at -sheets and fiber formation by the peptide R5 corresponding to the entire sequence of the R5 repeating unit of CsgA and R5T peptide, a troncated form of R5 that did not harbour the entire consensus sequence of CsgA. The formation of amyloid fibers was monitored by spectrophotometry in the presence of Congo red (congo red spectral shift assay), and fluorescence spectroscopy in the presence of thioflavin T, as previously described [9,10]. Both Congo red and Thioflavin T binding scan sometimes occur in the absence of amyloids, generating false positives. To confirm the results obtained with the Thioflavin T and Congo red assays, we performed imaging (polarized light microscopy, atomic force microscopy) to observe the concrete presence of fibers. Finally, the technique of infrared spectroscopy was used to characterize the secondary structure of protein assemblies. MATERIALS AND METHODS Synthetic Peptides The peptides R5 (133-SSVNVTQVGFGNNATTAHQY151) corresponding to the entire sequence of the R5 repeating unit of the CsgA curline protein and R5T (134SVNVTQVGFG-143) corresponding to a troncated form of the the R5 subunit were studied. R5T was selected through the TANGO software prediction of protein aggregation and formation of amyloid fibers [11]. This TANGO software calculates, for a protein sequence, a score of prediction of formation of aggregates rich in -sheets based on parameters © 2013 Bentham Science Publishers

2

Protein & Peptide Letters, 2013, Vol. 20, No. 8

from the protein sequence (presence of electrostatic charges, hydrophilic or hydrophobic residues, methylation sites), and from the nature of the solvent (presence of ions and their molarity, pH of the solution). For this in silico study, the chosen solvent was water at pH 7.4 with 200 mM salt. TANGO score obtained reflects the probability of fibers formation under the conditions requested. A TANGO score of 5 corresponds to a probability of 5% to form -aggregates for a selected peptide. TANGO scores were 1.56 and 11.36 for R5 and R5T, respectively. The peptides R5 and R5T were purchased from Proteogenix (Oberhausbergen, France). They were first dissolved in ultrapure water miliQ 18M at a concentration of 1 mM and further 10 times diluted in either of the 4 following buffers: Acetate / 20 mM acetic acid pH 4, acetate / acetic acid 20 mM pH 6, Tris / HCl 20 mM pH 7.4, carbonate / bicarbonate 20 mM pH 9. The incubation was performed at room temperature with a stirring of 80 revolutions per min. Thioflavin T Binding The supramolecular state of peptides R5 and R5T was first analysed by fluorescence spectroscopy in the presence of Thioflavin T. After 3 days of incubation in different buffers, 20 M of peptides were added to 20 M of thioflavin T (Sigma-Aldrich, Saint Quentin Fallavier, France) in 1 ml of Tris / HCl 20 mM pH 7.4. The fluorescence was measured with a spectrofluorimeter LS 50B (Perkin Elmer, Villebon sur Yvette, France). The excitation was carried out at 442 nm and emission was measured at 482 nm for a 1 cm optical path.

Lembré et al.

perature. Absorbance spectra were obtained with a spectrophotometer Uvicon (Secoman, Alès, France). The spectra were measured from 300 nm to 700 nm with an optical path of 1 cm. Polarized Light Microscopy The visualization of the fibers formed by R5 and R5T peptides was first done by polarized light in the presence of Congo red. A 3 days old peptide solution in ultra pure water was deposited on a glass slide previously cleaned by 2 sonication runs of 10 min in water 18M and mixed with equal volume of Congo red. The slides were dried at room temperature and the images were obtained with a microscope BX 60 (Olympus, Rungis, France) with objective X20. Atomic Force Microscopy For atomic force microscopy, a 3 days old peptide solution in ultra pure water was deposited on a glass slide previously cleaned and the slides were allowed to dry at room temperature. The images were obtained with a Dimension 3100 AFM controlled by Nanoscope IIIa (Veeco, Dourdan, France). Infrared Spectroscopy Peptides were dissolved in D2O to a final concentration of 20 mg/mL. After 3 days of incubation at room temperature with stirring, the infrared spectra were measured at a concentration of 6.6 mg/ml on a JASCO FTIR 6100 BIM (JASCO, Bouguenais, France) with an optical path of 100 nm.

Congo Red Binding The assemblies of peptides R5 and R5T were also analysed by absorbance spectroscopy in the presence of Congo red. For Congo red binding experiments, after incubation in buffer, 20μM peptide were added to 100 mM of Congo Red (Sigma-Aldrich, Saint Quentin Fallavier, France) in 20 mM Tris buffer at pH 7.4 and incubated for 1 hour at room tem-

RESULTS AND DISCUSSION In the presence of R5 or R5T, the fluorescence intensity of Thioflavin T was increased irrespective of the pH range in which the peptides were placed, Fig. (1). This increase in fluorescence was observed at 3 hours and was relatively stable after 3 days of incubation, Fig. (1). This result confirms

Figure 1. Fluorescence intensity at 482 nm of Thioflavin T in the presence of R5 or R5T peptide after 3 hours or 3 days of incubation in buffers at different pH. THT, Thioflavin T without peptide. a. u., arbitrary unit. The values are means of at least three independent experiments ± standard deviation.

Amyloid Fiber Formation by Synthetic Peptides

Protein & Peptide Letters, 2013, Vol. 20, No. 8

3

Figure 2. Absorbance spectrum of R5 and R5T peptides incubated 3 days in water in the presence of Congo red (CR). a. u., arbitrary unit.

previous results showing that the Thioflavin T fluorescence of CsgA over time follows a sigmoidal curve with distinguishable lag, growth, and stationary phases [5]. In this case, the Thioflavin T fluorescence signal has been shown to be unchanged after 8 h, remaining at approximately the same level for over 30 days [5]. The fluorescence signal was 2-3 times greater in the presence of R5T than for R5, suggesting that the type of the fibers formed by R5T may be different from R5. In the presence of R5 or R5T, we observed a red shift in the maximum absorbance of Congo red, revealing Congo red binding to fibers, Fig. (2). This confirmed the presence of amyloid fibers in the reaction mixture. The polarized light imaging allowed to highlight the Congo red binding on fiber peptides. In the presence of the peptide R5 or R5T, Congo red signal becomes birefringent demonstrating attachment onto amyloid fibers, Fig. (3A). Assemblies of peptides appear larger for R5 than for R5T. Atomic force imaging reveals that the two peptides formed rod-like fibers, Fig. (3B): the peptide R5 assembles into isolated and large-size agglomerated fibers; R5T fibers are well-ordered fibrillar structures sometimes combined as fagots but do not form aggregates of larger size. This can be related with the observation that the morphology of the fibers formed varies between the different repeat units of CsgA [5]. The morphology of R1 fibers is similar to that of those formed by purified CsgA, being generally longer than 1000 nm. R3 fibers are consistently shorter (ranging from 200 to 1000 nm) than those formed by CsgA. R5 fibers appear more rigid and aggregated than CsgA fibers. Hexapeptides QHGGGN and QFGGGN, which are present in R2 and R4, respectively, have also been shown to form well-ordered fibrillar structures ranging from 10 nm to 100 nm in diameter in vitro [8]: the morphology of QFGGGN fibers is close to R5T fibers whereas QHGGGN fibers look like R5 large aggregates. The presence of intermolecular -sheets in peptides preparations was analyzed by infrared spectroscopy. The presence of a band at 1622 cm-1 confirmed the presence of sheet in R5 and R5T, Fig. (4). The difference in intensity at 1622 cm-1 showed a greater number of intermolecular bonds for R5 than for R5T. Thus, the formation of a large number

of intermolecular bonds for R5 may conduct to the formation of big aggregates observed by polarized light in the presence of Congo red and by atomic force microscopy in this study and by transmission electron microscopy in the study of Wang et al. [5]. In addition to intermolecular -sheets (band at 1622 cm-1), the spectra of the peptides R5 and R5T also showed traces of trifluoroacetic acid used for peptide purification (band at 1672 cm-1), and non-organized structures (minor band at 1650 cm-1). Our results provide evidence that the entire consensus sequence SX5QXGX2NXAX3Q of the repeating units of CsgA is not essential for the formation of amyloid fibers in vitro. The 10 aminoacid long R5T peptide (134-SVNVTQV GFG-143) correponding to the X5QXGX2 sequence internal to the consensus sequence has self-assembly properties and forms amyloïd fibers in vitro at pH 4 to 9. By an alanine mutagenesis screen of the CsgA protein, Gln and Asn residues at positions 49, 54, 139 and 144 have been shown to be critical for curli assembly [12]. Remarkably, Gln residue at position 139 could not be replaced by Asn residues without interfering with curli assembly, suggesting that the side chain requirements were especially stringent at this position. The amyloidogenic R5T peptide contains the sequence region surrounding Gln at position 139 of the CsgA protein confirming the crucial role of this position for CsgA fibers assembly. In addition to Gln at position 139, the peptide R5 also contains the Asn residue at position 144 involved in side-chain contacts [12]. Thus, we hypothesize that R5 aggregation may involve more side-chain contacts than R5T assembly, conducting to the formation of aggregates of larger size. This is consistent with the results of infrared spectroscopy showing greater number of intermolecular bonds for R5 than for R5T (see above). In conclusion, a 10 amino acids peptide derived from the R5 sequence was sufficient for the spontaneous formation of amyloid fibrils but not to form large-size aggregates of fibers. The formation of large-size aggregates may be favored by an increase possibility to establish side-chain contacts and intermolecular bonds. Since the R5 domain contains sequences that contribute to CsgA binding to human proteins, the ability of shorter sequences derived from R5 to bind hu-

4

Protein & Peptide Letters, 2013, Vol. 20, No. 8

Lembré et al.

Figure 3. Images of polarized light microscopy in the presence of Congo red (A) and atomic force microscopy (B) of R5 and R5T peptides.

Figure 4. FTIR spectra of peptides R5 and R5T after 3 days of incubation.

man proteins is currently studied in our laboratory in order to identify the residues involved in adhesiveness of CsgA.

ACKNOWLEDGEMENTS Declared none.

CONFLICT OF INTEREST The authors confirm that this article content has no conflicts of interest.

REFERENCES [1]

Barrow, C.J.; Zagorski, M.G. Solution structures of beta peptide and its constituent fragments: relation to amyloid deposition. Science, 1991, 253, 179-182.

Amyloid Fiber Formation by Synthetic Peptides [2]

[3] [4]

[5]

[6] [7]

Protein & Peptide Letters, 2013, Vol. 20, No. 8

Larsen, P.; Nielsen, J.L.; Dueholm, M.S.; Wetzel, R.; Otzen, D.; Nielsen, P.H. Amyloid adhesions are abundant in natural biofilms. Environ. Microbiol., 2007, 9, 3077-3090. Barnhart, M.M.; Chapman, M.R. Curli biogenesis and function. Ann. Rev. Microbiol., 2006, 60, 131-147. Vidal, O.; Longin, R.; Prigent-Combaret, C.; Dorel, C.; Hooreman, M.; Lejeune, P. Isolation of an Escherichia coli K-12 mutant strain able to form biofilms on inert surfaces: involvement of a new ompR allele that increases curli expression. J. Bacteriol., 1998, 180, 2442-2449. Wang, X.; Smith, D.R.; Jones, J.W.; Chapman, M.R. In vitro polymerization of a functional Escherichia coli amyloid protein. J. Biol. Chem., 2007, 282, 3713-3719. Collinson, S.K.; Parker, J.M.; Hodges, R.S.; Kay, W.W. Structural predictions of AgfA, the insoluble fimbrial subunit of Salmonella thin aggregative fimbriae. J. Mol. Biol., 1999, 290, 741-56. Olsen, A.; Herwald, H.; Wikstrom, M.; Persson, K.; Mattsson, E.; Bjorck, L. Identification of two protein-binding and functional re-

Received: October 22, 2012

Revised: January 24, 2013

Accepted: January 24, 2013

[8]

[9]

[10] [11]

[12]

5

gions of curli, a surface organelle and virulence determinant of Escherichia coli. J. Biol. Chem., 2002, 277, 34568-34572. Cherny, I.; Rockah, L.; Levy-Nissenbaum, O.; Gophna, U.; Ron, E.O.; Gazit, E. The formation of Escherichia coli curli amyloid fibrils is mediated by prion-like peptide repeats. J. Mol. Biol., 2005, 352, 245-252. Klunk, W.E.; Jacob, R.F.; Mason, R.P. Quantifying amyloid by Congo red spectral shift assay. Methods Enzymol., 1999, 309, 285305. Nilsson, M.R. Techniques to study amyloid fibril formation in vitro. Methods, 2004, 34, 151-160. Fernandez-Escamilla, A.M.; Rousseau, F.; Schymkowitz, J.; Serrano, L. Prediction of sequence-dependent and mutational effects on the aggregation of peptides and proteins. Nat. Biotechnol., 2004, 22, 1302-1306. Wang, X.; Chapman, M.R. Curli provide the template for understanding controlled amyloid propagation. Prion, 2008, 2, 57-60.