Motility is a common behaviour in prokaryotes. Both bacteria and archaea use flagella for swimming motility, but it has been well documented that structures of the flagellum from these two domains of life are completely different, although they contribute to a similar function. Interestingly, information available to date has revealed that structurally archaeal flagella are more similar to bacterial type IV pili rather than to bacterial flagella. With the increasing genome sequence information and advancement in genetic tools for archaea, identification of the components involved in the assembly of the archaeal flagellum is possible. A subset of these components shows similarities to components from type IV pilus-assembly systems. Whereas the molecular players involved in assembly of the archaeal flagellum are being identified, the mechanics and dynamics of the assembly of the archaeal flagellum have yet to be established. Recent computational analysis in our laboratory has identified conserved highly charged loop regions within one of the core proteins of the flagellum, the membrane integral protein FlaJ, and predicted that these are involved in the interaction with the assembly ATPase FlaI. Interestingly, considerable variation was found among the loops of FlaJ from the two major subkingdoms of archaea, the Euryarchaeota and the Crenarchaeota. Understanding the assembly pathway and creating an interaction map of the molecular players in the archaeal flagellum will shed light on the details of the assembly and also the evolutionary relationship to the bacterial type IV pili-assembly systems.
In general, prokaryotes possess a variety of cellular structures on their surface, e.g. flagella and pili, that are thought to be involved in survival under various environmental conditions. Most of these surface structures are assembled in an orchestrated manner involving precise co-ordination of the components. Bacterial surface structures have been a topic of research for quite some time. Bacterial flagella and various pili are required for survival and host interaction during pathogenesis and have been extensively characterized [1–6]. In comparison, archaeal surface structures have come into focus quite recently [7–13]. Archaea represent the third domain of life. They have been isolated from a variety of differing environmental conditions and are known to thrive in different environmental niches at extreme pH values, temperatures and salinities [14–16]. Moreover, mesophilic archaea have recently been detected in marine waters, fresh water sediments and soil, thus confirming their occurrence in almost all types of ecosystem [17–19]. Studies on members from both subdivisions of archaea, i.e. Euryarchaeota and Crenarchaeota, revealed a variety of surface structures, including flagella, pili, hami and cannulae . Whereas in bacteria these structures are known to be involved in different kinds of motility as well as in the secretion of proteins and nucleic acids, not much is known about what purposes flagella and pili serve in archaea.
The flagellum is the best studied surface structure in archaea. Flagella from both subdomains, Euryarchaeota and Crenarchaeota, have been described [10,14,20]. Recent advances in genome sequencing have resulted in sequencing of many archaeal species from both subdomains. In silico analyses of all of these sequenced genomes were used to identify a genetic locus involved in flagellation in most of these archaeal species. Interestingly, components encoded by this genetic locus have no similarities to the components reported for bacterial flagella assembly. Instead, some of these archaeal flagella components were similar to T4P (type IV pilus) assembly and T2S (type II secretion) systems in Gram-negative bacteria. This finding has renewed interest among researchers working in microbial systems to examine more closely the structure, assembly and function of archaeal flagella.
Components of archaeal flagella-assembly systems
In archaea, the flagella locus contains several genes involved in proper assembly and function. The operon starts with genes encoding flagellins (flaA and/or flaB), the structural protein of the flagellum, followed either by fla-associated genes (flaC to flaJ) or by a distinct subset of these genes. Whereas most of the fla-associated genes are found in Euryarchaeaota, one or more of these genes are absent from the fla operon in Crenarchaeota (Figure 1). In general, flaC, flaD and flaE are only present in Euryarchaeota, such as Halobacterium or Methanococcus species [21–24]. In Halobacterium and Methanomicrobia, FlaC and FlaE are present as a single fused polypeptide . It has been shown by genetic analysis that deletions of flaCE or flaD from the fla gene cluster in Halobacterium salinarum resulted in mutant cells that possessed fewer flagella and had either very little (ΔflaD) or no motility (ΔflaCE) [24,25].
Model of the crenarchaeal and euryarchaeal flagella assembly systems
Recent findings in H. salinarum implicate a set of Che proteins (chemotaxis proteins, CheY, CheD and CheC2) in sensing environmental signals. This signal is transduced to FlaCE/D proteins, which are postulated to be components of the flagellar motor structure involved in directing the flagella to switch . Genomic analyses have revealed that genes encoding the Che proteins are in the vicinity of the fla operon, and protein–protein interaction analysis identified orphan interaction partners of the chemotaxis proteins CheY, CheD and CheC2 with the FlaCE/FlaD proteins. The systematic analysis in H. salinarum identified three new proteins (OE2401F, OE2402F and OE2403R) encoded by genes that are adjacent to the fla gene cluster. Two of these contain a DUF439 family protein, and the third one belongs to the HEAT_PBS family protein . These families of proteins have been identified in the chemotactic gene regions of all sequenced Haloarchaea, but not in any other archaeal genome .
Although the exact role of the other Fla proteins (FlaG, FlaF, FlaH, FlaI and FlaJ) is not understood properly, it has been shown that each of them is required for flagellation [7,24]. In crenarchaea, neither Che nor FlaCE/D proteins have been identified (Figure 1). Instead, the unknown gene flaXY is present in the fla operon, along with flaB, flaG, flaF, flaH, flaI and flaJ. Although no protein homologous with FlaXY could be identified, bioinformatics analysis revealed regions that are homologous with domains of methyl-accepting proteins, indicating a possible function in environmental signal sensing.
The assembly mechanism of archaeal flagella has been proposed to precede in a similar fashion as the mechanism for bacterial T4P [26–28]. T4P are involved in different processes such as twitching motility, DNA transfer and uptake, and pathogenicity .
At least three proteins in the archaeal flagella assembly system are homologous with components of T4P assembly systems [12,27]. These proteins are (i) the ATPase FlaI, which shows homology with PilB/PilT that are involved in assembly/disassembly of the pilus in T4P assembly systems, (ii) the polytopic membrane protein FlaJ, which is similar to inner membrane protein PilC in T4P, and (iii) the preflagellin peptidase PibD/FlaK, which resembles PilD, the prepilin peptidase in T4P assembly.
As in T4P assembly, the preflagellin peptidase FlaK/PibD processes the preflagellin by removing the positively charged N-terminal signal sequence before assembly [30–32]. Both FlaK from Methanococcus maripaludis and PibD from Sulfolobus solfataricus have been shown to have two important aspartic acid residues in their catalytic core, but, in contrast with PilD, they do not methylate the N-terminus of the mature flagellin during T4P assembly. Furthermore, processing of the signal sequence is essential, as flagella assembly does not occur in FlaK-deletion mutants [30,31]. Moreover, three-dimensional reconstruction analysis of electron microscopic images from the isolated archaeal flagella revealed a considerable degree of structural similarity between T4P and archaeal flagellum [13,33].
As described above, FlaI is a T2S and T4P assembly ATPase homologue, and FlaI from S. solfataricus has been shown to be capable of hydrolysing ATP in vitro . Detailed analysis of FlaI from Sulfolobus acidocaldarius revealed not only its ATP-hydrolysing capacity, but also its oligomerization upon binding of a non-hydrolysable ATP analogue in the presence of Mg(II) (A. Ghosh, unpublished work). These observations indicated a probable nucleotide-bound oligomeric conformation of FlaI similar to that already shown for the homologous T4P assembly counterparts, PilB/PilT [35,36]. In PilT, the disassembly ATPase, it was shown that the hexameric assembly switches between an open and a closed conformation depending on whether ADP or ATP respectively is bound. In the latter case, PilT regulates the disassembly process by hydrolysing the bound ATP in the T4P system .
Another interesting protein in the fla operon is FlaH, which has an incomplete ATPase-like domain as per in silico analysis. This potential ATPase-like protein contains only a typical Walker A motif and an incomplete Walker B motif. Similarly, the T4P assembly component PilT has an incomplete nucleotide-binding site, which nonetheless facilitates the disassembly process via ATP hydrolysis . For a long time, it has been postulated that FlaH interacts with FlaI and modulates its ATPase activity. Detailed biochemical analysis of FlaH in tandem with studies on the interactions of FlaI and FlaH would be an interesting topic of research.
FlaJ is a member of the PilC/GspF family and has been demonstrated to be important in the flagella-assembly process in H. salinarum, M. maripaludis and S. solfataricus [7,24,37]. This protein is predicted not only to contain seven to nine transmembrane domains and two large cytoplasmic loops, but also to form the central core complex in association with FlaI, on which the whole flagellum assembly takes place (Figure 1). Unfortunately, no biochemical data are presently available for this protein. Moreover, very little is known about the structures of T4P and T2S system homologues of PilC/GspF family proteins. It has been shown that these proteins (PilC/GspF) have apparently risen from an internal duplication of highly helical domains, which are thought to be located in either the cytoplasm or the periplasm according to evidence from multiple studies [27,38,39]. The crystal structure of the first cytoplasmic loop from the T2S system protein EspF (cyto-EspF56–171) from Vibrio cholerae showed a novel conformational helical fold and formed a dimer . A very recent report from the crystallization of the N-terminal cytoplasmic loop of inner membrane protein PilC (cyto1-PilC53–168) has confirmed further that a similar scenario is present in T4P assembly systems .
The conservation of such structural features across homologous systems implicates that FlaJ may serve as a platform for interaction with the other components in the assembly systems [40,41]. In silico analysis of FlaJ revealed the presence of two highly charged cytoplasmic loops, which are relatively well conserved among subdomains of archaea (Figure 2). Furthermore, analysis revealed that these two loops of S. acidocaldarius FlaJ are highly homologous with each other. Although these loops are believed to interact directly with the assembly ATPases, their specific function has yet to be elucidated.
Membrane topology of the polytopic membrane protein FlaJ in archaeal flagella assembly systems
Assembly and structural organization of the archaeal flagellum
Most of our current understanding of archaeal flagella and their assembly is the result of genetic analyses of various species of archaea. Genetic analysis of M. maripaludis has revealed that the deletion of FlaC, FlaF, FlaH, FlaG, FlaI and FlaJ resulted in non-motile non-flagellated cells, although the processed flagellins had been detected in the cell membrane . Although the cellular location of some of the flagellar components has been shown to be within the cell in M. voltae, cellular fractionation analysis revealed that components such as FlaB1/FlaB2, FlaC, FlaD, FlaE, FlaH and FlaI were localized in membrane fractions . Interestingly, whereas most of these proteins (FlaC, FlaD, FlaE and FlaH) were determined to contain no membrane-spanning domains, FlaI was predicted to have a membrane-spanning C-terminal domain. In crenarchaeal FlaI, however, no such membrane-spanning domain has been detected. Moreover, FlaI was found to be present in both the cytosol and the membrane in cellular fractionating analysis of S. acidocaldarius (A. Ghosh, unpublished work). Subcellular localization studies have also indicated a possible interaction between FlaI and FlaJ (A. Ghosh, unpublished work).
Structural analyses of euryarchaeal and crenarchaeal flagella have revealed that archaeal flagella are thin right-handed helical filaments with a diameter ranges from 11 to 14 nm [9,12,33]. In general, more than one subunit of flagellin proteins (flaA and flaB) are found in archaea. It has been postulated that these different flagellin subunits contribute to variable structural symmetries using differential subunit packing in the growing flagellum. In the euryarchaeon Haloarcula marismortui, a recent study has shown that different flagellins are involved in flagellin switching, which might protect cells from hypothetical flagellum-specific haloarchaeal phages . Interestingly, in comparison with the other archaeal species, only a single flagellin (flaB) gene has been detected in Sulfolobales [10,43]. However, despite the presence of a variable number of flagellin subunits, it has recently been shown that both Sulfolobus and Halobacterium have a common subunit packing arrangement in the flagellum [9,13,44].
The packing arrangement of the flagellins in the archaeal flagellum is completely different from that of bacterial flagella [9,13,33]. This observation is indicative of the distant evolutionary segregation that exists between bacterial and archaeal flagella. The most intriguing difference denoted during structural analysis of bacterial and archaeal flagella was the absence of a central channel in archaeal flagella, which was first observed some time ago in bacterial flagella, where it is known to be important for flagellar assembly . It has been postulated that archaeal flagellum assembly takes place by adding subunits to the base of the growing flagellum on the cell surface, on the basis of the aforementioned similarities to T4P assembly.
Despite its structural resemblance to a T4P, the archaeal flagellum has been shown to be more similar to a bacterial flagellum with regard to its rotation and involvement in surface attachment [45–47].
In H. salinarum, the capacity to swim involves the rotation of polymorphic flagellar bundles, as has been shown by cellular responses to phototactic pulses . Moreover, it has been demonstrated that the haloarchaeal flagellar motor is not driven by the PMF (protonmotive force) as in bacteria, but is dependent directly on ATP hydrolysis or indirectly using ATP-dependent ion gradient .
Archaeal flagella are involved in multiple processes
In both H. salinarum and M. voltae, motility has been shown to be dependent on flagella [3,48–50]. In the thermoacidophilic crenarchaeon S. solfataricus, flagellum-dependent motility has been confirmed by combining genetic analysis with motility assays . In S. acidocaldarius, temperature-dependent motility has been described in a quantitative manner . Recent studies have revealed that archaeal flagella are involved not only in motility, but also in adherence to different surfaces. S. solfataricus flagella-deletion mutants were unable to attach to surfaces such as glass, mica, carbon-coated grids and pyrite . Despite being a prerequisite for the initial attachment to a surface, the flagella did not play an important role in the manifestation of biofilms formed by S. solfataricus . Pyrococcus furiosus cells adhered to mica and carbon-coated grids and formed a cable-like network made up of flagella filaments .
Colonization of P. furiosus cells on glass surfaces could not take place directly and only occurred when the surface was first colonized by Methanopyrus kandleri cells. By using their flagella, P. furiosus established direct cell–cell contact with M. kandleri cells that were already adhering to the glass surface . In contrast with this, recent studies in Haloferax volcanii have shown flagellum-independent surface adhesion, although the flagella were shown to be involved in motility . Interestingly, a PibD-deletion mutant of H. volcanii has recently been shown to be non-motile, but nevertheless had the capacity to adhere to surfaces such as glass . In H. volcanii, a distinct structure resembling a T4P has been postulated to have an involvement in adherence to surfaces ; however, in order to unravel the paradox behind the structure and function of archaeal flagella, detailed molecular analysis of the assembly process and the mode of action will be essential.
Despite the limited amount of details presently available regarding the structure and assembly of archaeal flagella, it has become increasingly evident from multiple studies that flagella play important roles in a variety of cellular processes in archaea [3,12,47,51,53,55,56]. Moreover, core components of the T4P and the archaeal flagella-assembly systems are significantly similar to one another, thus raising intriguing questions as to the evolutionary relationship between these two systems. In the T4P-assembly system, limited biochemical evidence exists for piecing together the process of flagellar assembly as well as the detailed interactions between assembly components. Therefore any structural evidence or interaction map of the components involved in the flagellar assembly process will not only improve our understanding of the archaeal flagellum assembly, but also shed light on T4P assembly. Furthermore, owing to the absence of an outer membrane in archaea, a more simplified version of the T4P-assembly machinery must be present that will facilitate our understanding of the system.
Molecular Biology of Archaea II: A Biochemical Society Focused Meeting held at Robinson College, Cambridge, U.K., 16–18 August 2010. Organized and Edited by Stephen Bell (Oxford, U.K.) and Finn Werner (University College London, U.K.).
A.G. received a Max Planck postdoctoral fellowship and S.-V.A. was supported by a VIDI grant from the Dutch Science Organization (NWO) and intramural funds of the Max Planck Society.