Contribution of holins to protein trafficking: secretion, leakage or lysis?

Contribution of holins to protein trafficking: secretion, leakage or lysis?

Letter Contribution of holins to protein trafficking: secretion, leakage or lysis? Mickae¨l Desvaux INRA, UR454 Microbiology, Clermont-Ferrand Resear...

623KB Sizes 0 Downloads 3 Views

Letter

Contribution of holins to protein trafficking: secretion, leakage or lysis? Mickae¨l Desvaux INRA, UR454 Microbiology, Clermont-Ferrand Research Centre, F-63122 Saint-Gene`s Champanelle, France

Holins are xenologues of phagic origin virtually present in all bacteria. Primarily responsible for some cytocidal phenotypes (Box 1) [1], holins are also presented as an alternative secretion system for some proteins without a signal peptide (SP) [2,3]. This export system is considered specific to prophage-related murein hydrolases, i.e. endolysins, but also to some important human virulence factors, such as Stx1 (Shiga-like toxin 1), SheA (silent hemolysin A) or TcdAB (Clostridium difficile toxins A and B). However, a recent article by Olling et al. throws a ‘pave´ dans la mare’ by challenging the notion of holins being involved in the secretion of bacterial proteins [4]. Indeed, it demonstrates that the release of TcdA and TcdB does not depend on the holin TcdE (TC#1.E.19; Transporter Classification Database: http://www.tcdb.org/), which could only be synthesized as an antiholin (Box 1). This finding is of great interest not only because it refutes some original findings but also because TcdE is often presented as a paradigm for the secretion of proteins other than endolysins via holins [3]. More generally, it revives the matter of the actual contribution of holins to protein trafficking in bacteria. In other words, can holins be truly regarded as protein secretion systems? Holins are often depicted as specific to their cognate endolysins, where they control their transport and even their activity [5]. Endolysins can either (i) possess a Sec-dependent SP, which is uncleaved and serves as a membrane signal anchor until the holin triggering de-energizes the membrane to consequently release and activate them or (ii) lack a SP, in which case the way holin mediates their passage across the membrane is unclear. The first situation involves pinholins, e.g. S21, which generate small pores (diameter <2 nm), the second situation involves canonical holins, e.g. S105, forming large pores of 340 nm diameter on average but also some above 1 mm [6]. Interestingly, only one pore from S105 is present in most bacterial cells, suggesting it grows rapidly from a single nucleating point. Considering that secretion like export implies active transport [2], holins are unlikely to promote transport driven by ATP hydrolysis or by the proton-motive force, such as the Sec or Tat systems, respectively. Upon holin triggering, conformational modifications concur with pore formation (Box 1). In the type V secretion system (T5SS), the translocation driving force would result from free energy released upon structural rearrangements [7]. Contrary to pinholin, however, direct interaction between the other types of holin and endolysin for folding has never been reported. Unless an (unlikely)

interaction with other membrane proteins supplying energy for membrane transport occurs, holins are probably not protein translocation systems per se (Box 1). So what are the alternatives to a secretion mechanism? The current model proposes holins provide a pore for passive and nonspecific transport [5,8], in other words the cytoplasmic components would leak through the holes, sweeping away the endolysin (Box 1). Once the holes by S105 are formed, cell lysis occurs within seconds [9]. Endolysin activity was demonstrated to accelerate the cytolethal event. Using an endolysin-b-galactosidase protein fusion, accessibility and activity against the cell wall was demonstrated suggesting the membrane lesion was not specific but sufficient to accommodate fully folded proteins as large as 450 kDa [10]. In line with massive membrane disruption [6,8], however, another interpretation is not leakage through the pore but release of the whole cytoplasmic content at once following the loss of membrane integrity as the end result of the formation of a micron-scale hole (Box 1); accumulated endolysins would then have access to the cell wall to complete the cell lysis. In both Gram-negative and Gram-positive bacteria, holins allow their cognate endolysins to access the cell wall. However, additional proteins, the spanins, are required to complete cell lysis by disrupting the outer membrane in Gram-negative bacteria. This is a major difference with consequences on protein trafficking [2]. It must be stressed that the current model is based on bacteriophage holins (Box 1), essentially from investigations on S holin, ultimately responsible for the release of virions in the course of the lytic cycle. While other holin families rely on this model, the actual molecular mechanisms involved in the release of their cognate endolysins remain unknown. For instance, the specificity, size, distribution or resilience of the pores of the CidA holin involved in programmed cell death (PCD) await investigation [5]. It must be pinpointed whether holin involved in phage infection cycle or in PCD is a different matter, e.g. the timing

Box 1. Holins in a nutshell Holins form a homo-oligomeric pore in the cytoplasmic membrane [5]. Whereas some holins feature the release of virions, such as the prototypical S holin from phage l, others are involved in programmed cell death (PCD), e.g. the CidA holin. Holins (TC#1.E) are a diverse group of proteins falling into 23 distinct families (Transporter Classification Database: http://www.tcdb.org/). The prime function of holins is the access of their associated endolysins to the cell wall (Figure I).

Corresponding author: Desvaux, M. ([email protected]).

259

Letter

Trends in Microbiology June 2012, Vol. 20, No. 6

(a)

N Class I

C

Class II

EM/PP CM N

Ex: S21 holin

C

(b)

C

N

N

Ex: T4 holin

CP

5

Ex: CidA holin

S1 0

S1 0

7

Ex: S105 holin

C

S

R

PR'

(c)

EM/PP CM CP S105 holin

(i)

(ii)

(iii)

Endolysin R

(d)

CM

CM

CM

(i)

(ii)

(iii)

Key: : Holin pore

: Cognate endolysin

: Energy for transport

: Other cytoplasmic components

TRENDS in Microbiology

Figure I. Structure, function and mechanism of holins. (a) Holins are generally distributed between class I and class II, exhibiting three and two transmembrane domains (TMDs), respectively; however, some holins have more or less TMDs. Topology is either based on prediction tools or biochemical and genetic evidence. (b) As a typical holin–endolysin system, S and R genes (encoding the holin and endolysin, respectively) are found in tandem and encoded in an operon. S encodes both the S105 holin and its inhibitor, the S107 antiholin, as a result of dual-start sites. By forming abortive dimers with S105, the proportion of S107 is a key control mechanism for timing of lysis. However, once the first lesion is formed, the membrane potential collapses, eliminating the inhibitory capacity of S107. (c) The successive events triggering holin are: (i) the accumulation of mobile S105 dimers within the membrane; (ii) the raft nucleation at a critical concentration causing the collapse of membrane energization; which in turn (iii) leads to conformational changes and pore formation [8]. The current model proposes the cytoplasmic components leak through the pore. (d) The three theoretical models of endolysin access to cell wall following holin triggering include: (i) specific and active translocation of endolysins through a pore; (ii) passive diffusion of cytoplasmic components through a hole, including accumulated endolysins; or (iii) loss of membrane integrity due to the formation of a growing micron-scale lesion resulting in the release of the whole cytoplasmic contents. Abbreviations: CP, cytoplasm; CM, cytoplasmic membrane; EM, extracellular milieu; PP, periplasm.

260

Letter issue is more relevant to the former (Box 1). Considering the diversity of holin families, in terms of sequence and topology, it cannot be excluded that each of them exhibits some particularity. This stresses the need for further investigation, especially at late stages of holin triggering. Could some interactions between holin and a cognate protein (not only endolysin) occur and provide the driving force for secretion? Do holins and antiholins integrate in a YidC-dependent manner, i.e. via the membrane insertase/ integrase transferring the protein hydrophobic regions to the cytoplasmic membrane? Does holin triggering lead to the rapid loss of membrane integrity resulting in the release of the cytoplasmic contents or to local membrane disruption through which the cytoplasmic components seep out? Could antiholin interaction with holin be involved in formation of transitory pores, which could lead to leakage but not to systematic cell lysis? Along these lines, the actual contribution of holins to nonclassical secretion would deserve further attention [11]. Indeed, the presence of primarily cytoplasmic proteins in the extracellular milieu remains a puzzling phenomenon [12]. Although it is independent from cell lysis, knowing these proteins accumulate in the cytoplasm before being located extracellularly has a particular resonance with holin triggering. As a protein family first discovered more than 20 years ago and regrouping such different protein classes, time has come to investigate this diversity. Beyond bacteriophage S holin and the Cid/Lrg PCD system, studies on the molecular functions and mechanisms of other holins are scarce or remain to be performed. The odds are that some specificity will be uncovered that could further deepen our comprehension of the current

Trends in Microbiology June 2012, Vol. 20, No. 6

models, especially regarding the precise contribution to protein trafficking in bacteria. References 1 Bayles, K.W. (2003) Are the molecular strategies that control apoptosis conserved in bacteria? Trends Microbiol. 11, 306–311 2 Desvaux, M. et al. (2009) Secretion and subcellular localizations of bacterial proteins: a semantic awareness issue. Trends Microbiol. 17, 139–145 3 Tjalsma, H. et al. (2004) Proteomics of protein secretion by Bacillus subtilis: separating the ‘‘secrets’’ of the secretome. Microbiol. Mol. Biol. Rev. 68, 207–233 4 Olling, A. et al. (2012) Release of TcdA and TcdB from Clostridium difficile cdi 630 is not affected by functional inactivation of the tcdE gene. Microbiol. Pathog. 52, 92–100 5 Rice, K.C. and Bayles, K.W. (2008) Molecular control of bacterial death and lysis. Microbiol. Mol. Biol. Rev. 72, 85–109 6 Dewey, J.S. et al. (2010) Micron-scale holes terminate the phage infection cycle. Proc. Natl. Acad. Sci. U.S.A. 107, 2219–2223 7 Junker, M. et al. (2009) Vectorial transport and folding of an autotransporter virulence protein during outer membrane secretion. Mol. Microbiol. 71, 1323–1332 8 White, R. et al. (2011) Holin triggering in real time. Proc. Natl. Acad. Sci. U.S.A. 108, 798–803 9 Gru¨ndling, A. et al. (2001) Holins kill without warning. Proc. Natl. Acad. Sci. U.S.A. 98, 9348–9352 10 Wang, I.N. et al. (2003) Sizing the holin lesion with an endolysinb-galactosidase fusion. J. Bacteriol. 185, 779–787 11 Walker, S.A. and Klaenhammer, T.R. (2001) Leaky Lactococcus cultures that externalize enzymes and antigens independently of culture lysis and secretion and export pathways. Appl. Environ. Microbiol. 67, 251–259 12 Yang, C.K. et al. (2011) Nonclassical protein secretion by Bacillus subtilis in the stationary phase is not due to cell lysis. J. Bacteriol. 193, 5607–5615 0966-842X/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tim.2012.03.008 Trends in Microbiology, June 2012, Vol. 20, No. 6

261