Intrinsic aggregation propensity of the CsgB nucleator protein is crucial for curli fiber formation

https://doi.org/10.1016/j.jsb.2016.05.012Get rights and content

Abstract

Several organisms exploit the extraordinary physical properties of amyloid fibrils forming natural protective amyloids, in an effort to support complex biological functions. Curli amyloid fibers are a major component of mature biofilms, which are produced by many Enterobacteriaceae species and are responsible, among other functions, for the initial adhesion of bacteria to surfaces or cells. The main axis of curli fibers is formed by a major structural subunit, known as CsgA. CsgA self-assembly is promoted by oligomeric nuclei formed by a minor curli subunit, known as the CsgB nucleator protein. Here, by implementing AMYLPRED2, a consensus prediction method for the identification of ‘aggregation-prone’ regions in protein sequences, developed in our laboratory, we have successfully identified potent amyloidogenic regions of the CsgB subunit. Peptide-analogues corresponding to the predicted ‘aggregation-prone’ segments of CsgB were chemically synthesized and studied, utilizing several biophysical techniques. Our experimental data indicate that these peptides self-assemble in solution, forming fibrils with characteristic amyloidogenic properties. Using comparative modeling techniques, we have developed three-dimensional models of both CsgA and CsgB subunits. Structural analysis revealed that the identified ‘aggregation-prone’ segments may promote gradual polymerization of CsgB. Briefly, our results indicate that the intrinsic self-aggregation propensity of the CsgB subunit, most probably has a pivotal role in initiating the formation of curli amyloid fibers by promoting the self-assembly process of the CsgB nucleator protein.

Introduction

Amyloids are an after-effect of deposition of ordered protein fibrillar arrangements, known as amyloid fibrils (Chiti and Dobson, 2006). Amyloid fibrils are formed by otherwise soluble proteins or peptides that convert under certain conditions into insoluble fibrous aggregates (Dobson, 1999). Impressively, several proteins with important but otherwise unrelated functions have been associated with amyloid deposition, although they have neither sequence nor structural apparent similarities (Sipe et al., 2014). A large number of widespread diseases, such as AL (Amyloid Light-chain), AA (Amyloid A) or ATTR (Transthyretin-related) amyloidosis, neurodegenerative diseases (Alzheimer’s, Parkinson’s and Creutzfeldt-Jakob’s disease among others), type II diabetes and many more are a consequence of unrestrained deposition of amyloid causing tissue damage and degeneration (Sipe et al., 2014). In contrast, accumulated evidence has shown that occasionally organisms spanning from bacteria to humans exploit the extraordinary intrinsic properties of amyloids in order to support fundamental physiological processes (Iconomidou and Hamodrakas, 2008, Shewmaker et al., 2011). Typical examples of functional amyloid include the extracellular protective coats of several organisms (Iconomidou et al., 2000, Louros et al., 2013, Louros et al., 2014a, Louros et al., 2016b), the intracellular Pmel17 template which is utterly important for the biosynthesis of melanin (Fowler et al., 2006, Louros and Iconomidou, 2015, Louros et al., 2016a) and the formation of biofilms by gram negative bacteria (Hammar et al., 1995).

Bacteria are able to survive and colonize in a diverse variety of environments. Key feature to their resilience is their ability to grow in colonies and produce a complex matrix of extracellular polymeric substances (Donlan, 2002). Using this matrix, bacteria are able to sculpt three-dimensional structures, called biofilms, which shelter the inhabitants from environmental stress (Donlan, 2002). In the case of many Enterobacteriaceae such as Escherichia coli or Salmonella enterica, the major proteinaceous component of this matrix consists of fibers called curli that are involved in cell-surface and cell-cell contacts and adhesion (Barnhart and Chapman, 2006). Curli fibers exhibit characteristic properties of amyloid fibrils (Chapman et al., 2002) and are composed of two basic proteins; namely, a major structural subunit, known in E. coli as CsgA, and a minor subunit serving also as a nucleator protein, known as CsgB, both encoded by a common operon (Hammar et al., 1996, Hammar et al., 1995). The two proteins are of identical size (151aa) and comprise similar structural segments (Hammar et al., 1996). Specifically, CsgA and CsgB are built up of three regions, namely a signal peptide, an N-terminal chaperone-binding peptide segment and five consecutive repeat subunits, composed of 22–23 residues, which constitute the amyloid core of curli fibers (Hammer et al., 2007, Wang and Chapman, 2008). Each repeat is composed of a strand-loop-strand motif predicted to form two parallel β-sheets (Wang and Chapman, 2008, Wang et al., 2008).

Both proteins are capable of forming fibers in vitro, however detailed evidence indicates that CsgB forms fibers with a significantly faster rate (Hammar et al., 1996). Additionally, the aggregation kinetics of CsgA is accelerated in the presence of CsgB (Hammer et al., 2012). However, in vivo curli fibers can only be formed in the presence of both subunits (Barnhart and Chapman, 2006, Blanco et al., 2012, Hammer et al., 2012, Wang et al., 2007), accumulated at ratios approximately 20:1 (CsgA:CsgB) (Van Gerven et al., 2015, White et al., 2001). Furthermore, detailed immunoelectron microscopy studies have shown that CsgB can form short polymers on the cell surface in the absence of CsgA (Bian and Normark, 1997). Finally, CsgB has been proposed to be responsible for creating the initial oligomers/nuclei acting as templates and inducing the polymerization of CsgA, which self-assembles forming the major axis of the curli fiber (Hammer et al., 2007, Shu et al., 2012).

Experimental and theoretical evidence has indicated that amyloid formation is induced by specific short sequence regions of a polypeptide chain that are prone to aggregation, hence regulating the overall aggregation tendency of the protein (Lopez de la Paz and Serrano, 2004, Louros et al., 2015a, Louros et al., 2015b, Teng et al., 2012). In this work, we have identified six individual aggregation-potent regions of the minor curli subunit, by comparing the amyloidogenic profile of both CsgA and CsgB curli proteins of E. coli. Structural studies of peptide-analogues corresponding to the identified regions indicate their ability to self-assemble forming fibrils with characteristic amyloidogenic features. Our findings suggest that the ‘aggregation-prone’ segments could probably be responsible for the ability of CsgB to self-assemble in faster rates than CsgA, in vitro, and may also promote the nucleation capabilities of the former. Finally, based on our results, we propose a possible mechanism for both the in vivo and in vitro curli fiber self-assembly process, by attempting to shed some light on the vague and complicated nucleation process of curli fibers.

Section snippets

Aggregation propensity sequence analysis of CsgA and CsgB

Protein sequences of CsgA and CsgB from E. coli were obtained from Uniprot (Accession Numbers: P28307 and P0ABK7). Initially, the Sec-signal peptide and the 22-residue long N-terminal domain that precedes 5 imperfect repeats in both CsgA and CsgB (Fig. S1) were removed. In order to track down the aggregation propensity of both curli subunits, we implemented AMYLPRED2, a consensus aggregation propensity prediction tool, which was developed by our lab (Tsolis et al., 2013), on the remaining

CsgB exhibits increased aggregation propensity compared to CsgA

The aggregation potency of both the major and minor subunit of E. coli curli fibers was calculated and analyzed utilizing AMYLPRED2. Impressively, although the aforementioned sequences share relatively high sequence homology, significant differences emerge regarding their aggregation tendency. As the results of AMYLPRED2 clearly indicate, severely lower aggregation potency was predicted for CsgA in comparison to CsgB, since the former only presents segments with an extraneous minor aggregation

Funding

The research leading to these results has received partial funding from the European Community’s Seventh Framework Programme (FP7/2007-2013) under BioStruct-X (grant agreement N°283570). The infrastructure that the authors used for the present work is associated with the Greek National Research Infrastructure in Structural Biology, Instruct-EL.

Acknowledgements

We thank the Institute of Biology, Medicinal Chemistry and Biotechnology at National Hellenic Research Foundation for hospitality. We acknowledge the help of Dr. Evangelia Chrysina with the X-ray diffraction experiments. The help of Dr. George Baltatzis and Prof. Efstratios Patsouris and the use of the Morgagni Microscope at the 1st Department of Pathology, Medical School, University of Athens are also gratefully acknowledged. We also thank the University of Athens for support. Finally, the

References (86)

  • S. Krimm et al.

    Vibrational spectroscopy and conformation of peptides, polypeptides, and proteins

    Adv. Protein Chem.

    (1986)
  • N.N. Louros et al.

    An N-terminal pro-atrial natriuretic peptide (NT-proANP) ‘aggregation-prone’ segment involved in isolated atrial amyloidosis

    FEBS Lett.

    (2014)
  • N.N. Louros et al.

    Chameleon ‘aggregation-prone’ segments of apoA-I: a model of amyloid fibrils formed in apoA-I amyloidosis

    Int. J. Biol. Macromol.

    (2015)
  • F. Shewmaker et al.

    Structural insights into functional and pathological amyloid

    J. Biol. Chem.

    (2011)
  • R. Singh et al.

    Biofilms: implications in bioremediation

    Trends Microbiol.

    (2006)
  • M. Sunde et al.

    Common core structure of amyloid fibrils by synchrotron X-ray diffraction

    J. Mol. Biol.

    (1997)
  • N. Van Gerven et al.

    Bacterial amyloid formation: structural insights into curli biogenesis

    Trends Microbiol.

    (2015)
  • G. Vriend

    WHAT IF: a molecular modeling and drug design program

    J. Mol. Graph. Model.

    (1990)
  • X. Wang et al.

    Sequence determinants of bacterial amyloid formation

    J. Mol. Biol.

    (2008)
  • X. Wang et al.

    In vitro polymerization of a functional Escherichia coli amyloid protein

    J. Biol. Chem.

    (2007)
  • X. Wang et al.

    The molecular basis of functional bacterial amyloid polymerization and nucleation

    J. Biol. Chem.

    (2008)
  • A.P. White et al.

    Structure and characterization of AgfB from Salmonella enteritidis thin aggregative fimbriae

    J. Mol. Biol.

    (2001)
  • L. Zhang et al.

    Biofilm formation and electricity generation of a microbial fuel cell started up under different external resistances

    J. Power Sources

    (2011)
  • M.M. Barnhart et al.

    Curli biogenesis and function

    Annu. Rev. Microbiol.

    (2006)
  • Z. Bian et al.

    Nucleator function of CsgB for the assembly of adhesive surface organelles in Escherichia coli

    EMBO J.

    (1997)
  • A.T. Brunger et al.

    Crystallography & NMR system: a new software suite for macromolecular structure determination

    Acta Crystallogr. D Biol. Crystallogr.

    (1998)
  • A.W. Bryan et al.

    Structure-based prediction reveals capping motifs that inhibit beta-helix aggregation

    Proc. Natl. Acad. Sci. U.S.A.

    (2011)
  • J.D. Bryers

    Medical biofilms

    Biotechnol. Bioeng.

    (2008)
  • B. Cao et al.

    Structure of the nonameric bacterial amyloid secretion channel

    Proc. Natl. Acad. Sci. U.S.A.

    (2014)
  • M.R. Chapman et al.

    Role of Escherichia coli curli operons in directing amyloid fiber formation

    Science

    (2002)
  • F. Chiti et al.

    Protein misfolding, functional amyloid, and human disease

    Annu. Rev. Biochem.

    (2006)
  • O. Conchillo-Sole et al.

    AGGRESCAN: a server for the prediction and evaluation of “hot spots” of aggregation in polypeptides

    BMC Bioinformatics

    (2007)
  • E.A. Cooper et al.

    Fourier transform infrared spectroscopy investigations of protein structure

    Pharm. Biotechnol.

    (1995)
  • S.J. de Vries et al.

    The HADDOCK web server for data-driven biomolecular docking

    Nat. Protoc.

    (2010)
  • D. Divry et al.

    Sur les proprietes optiques de l’amyloide

    C.R. Soc. Biol.

    (1927)
  • R.M. Donlan

    Biofilms: microbial life on surfaces

    Emerg. Infect. Dis.

    (2002)
  • D.T. Downing et al.

    Molecular modelling indicates that the pathological conformations of prion proteins might be beta-helical

    Biochem. J.

    (1999)
  • S.J. Edwards et al.

    Applications of biofilms in bioremediation and biotransformation of persistent organic pollutants, pharmaceuticals/personal care products, and heavy metals

    Appl. Microbiol. Biotechnol.

    (2013)
  • N. Eswar et al.

    Comparative protein structure modeling using Modeller

    Curr. Protoc. Bioinformatics

    (2006)
  • M.L. Evans et al.

    Curli biogenesis: order out of disorder

    Biochim. Biophys. Acta

    (2013)
  • A.M. Fernandez-Escamilla et al.

    Prediction of sequence-dependent and mutational effects on the aggregation of peptides and proteins

    Nat. Biotechnol.

    (2004)
  • D.M. Fowler et al.

    Functional amyloid formation within mammalian tissue

    PLoS Biol.

    (2006)
  • O.V. Galzitskaya et al.

    Prediction of amyloidogenic and disordered regions in protein chains

    PLoS Comput. Biol.

    (2006)
  • Cited by (0)

    View full text