The type IV pilin, PilA, is required for full virulence of Francisella tularensis subspecies tularensis
© Forslund et al; licensee BioMed Central Ltd. 2010
Received: 18 December 2009
Accepted: 26 August 2010
Published: 26 August 2010
All four Francisella tularensis subspecies possess gene clusters with potential to express type IV pili (Tfp). These clusters include putative pilin genes, as well as pilB, pilC and pilQ, required for secretion and assembly of Tfp. A hallmark of Tfp is the ability to retract the pilus upon surface contact, a property mediated by the ATPase PilT. Interestingly, out of the two major human pathogenic subspecies only the highly virulent type A strains have a functional pilT gene.
In a previous study, we were able to show that one pilin gene, pilA, was essential for virulence of a type B strain in a mouse infection model. In this work we have examined the role of several Tfp genes in the virulence of the pathogenic type A strain SCHU S4. pilA, pilC, pilQ, and pilT were mutated by in-frame deletion mutagenesis. Interestingly, when mice were infected with a mixture of each mutant strain and the wild-type strain, the pilA, pilC and pilQ mutants were out-competed, while the pilT mutant was equally competitive as the wild-type.
This suggests that expression and surface localisation of PilA contribute to virulence in the highly virulent type A strain, while PilT was dispensable for virulence in the mouse infection model.
Francisella tularensis is a Gram-negative facultative intracellular bacterial pathogen and the causative agent of tularemia. Infections have been reported in a range of vertebrates as well as invertebrates . Natural infections can occur via ingestion of contaminated food or water, handling of infected animals, bites by infected arthropods, including mosquitoes and ticks, or via inhalation . F. tularensis is divided into four subspecies, where ssp. holarctica (type B) is most widely spread and found in the major part of Europe, Asia, and North America. F. tularensis ssp. tularensis (type A) is found exclusively in North America and ssp. mediasiatica in Central Asia. Finally, ssp. novicida has been isolated in several locations in North America, as well as in Australia [3, 4]. Human infections are mainly caused by type A or type B strains, where type A strains are significantly more virulent than type B strains. Our knowledge regarding virulence determinants in F. tularensis is rather limited. However, available genome information [5, 6] together with development of genetic tools , has resulted in increased understanding of the molecular mechanisms of F. tularensis infections.
The genome of F. tularensis encodes gene clusters involved in secretion and assembly of type IV pili (Tfp) . Tfp are complex adhesins involved in important host cell interactions for human pathogens like Neisseria spp., Pseudomonas aeruginosa and Vibrio cholerae[8–11]. The pilus fiber is composed of one major pilin subunit and several additional minor pilins required for function and/or assembly of the pilus [12, 13]. However, the exact roles of the minor pilins are still not completely understood. The pilus is translocated to the cell surface via the secretin, PilQ, which forms a pore in the outer membrane through which the pilus is transported and extended . PilD is a peptidase cleaving the prepilin subunits  and PilC is a transmembrane protein spanning across the plasma membrane . Furthermore, two ATPases, PilB and PilT, are involved in extension and retraction, of the pilus [16, 17]. In some bacteria Tfp can mediate twitching motility, an activity that is PilT dependent .
There is evidence that F. tularensis expresses Tfp-like surface structures on the bacterial surface [19–21], and the putative pilin, PilA, has been shown to be required for virulence of type B strains in a mouse infection model . Interestingly, due to direct repeat mediated deletion, the pilA gene has been lost in the attenuated live vaccine strain LVS [22, 23], supporting the significance of PilA for virulence . There are also other potentially significant differences between different F. tularensis subspecies. In ssp. novicida that is non-pathogenic for humans, PilA differs in the amino acid sequence compared to the virulent type A strain SCHU S4 . On the contrary, pilA of virulent type B strains is essentially identical to the corresponding gene in type A strains, however, several other differences are apparent between the two subspecies. Two predicted pilin genes, pilE and pilV, and the ATPase encoding gene, pilT, are pseudogenes in type B strains [19, 21, 22, 26]. These differences are likely to be significant since pilA has been linked to virulence of type B strains and is missing in the attenuated vaccine strain LVS. Furthermore, PilA of ssp. novicida was recently shown to be involved in protein secretion that was coupled to Tfp [20, 25]. Interestingly, mutation of pilA and loss of protein secretion resulted in increased virulence in a mouse infection model . As the human pathogenic type A and type B strains do not secrete detectable levels of proteins in vitro, it is possible that one step in the evolution of human pathogenic variants of F. tularensis from ssp. novicida has involved loss of protein secretion as a consequence of changes in PilA structure and function.
In this work we wanted to address the question if PilA is involved in virulence of the highly pathogenic type A strain SCHU S4, similarly to what we have previously shown for type B strains, and if Tfp secretion and assembly genes are required for virulence.
Construction of non-polar pilin gene mutants
In a recent study, we were able to demonstrate that the pilA gene can be lost by a deletion event mediated by direct repeats flanking the gene . Type B strains lacking pilA were found to be attenuated for virulence in a mouse infection model. In this study we wanted to extend this work to the highly pathogenic type A strain SCHU S4, and therefore we constructed a specific pilA deletion mutant using our previously described allelic exchange technique . In addition, to address the significance of secretion and assembly of PilA, we also engineered in-frame deletions in pilC and pilQ, encoding a transmembrane protein and a secretin, respectively. For some pathogens, Tfp expression is associated with a unique ability to retract the pili, a phenotype depending on the ATPase PilT. Interestingly, pilT appears to be functional in type A strains, while it is a pseudogene in the less pathogenic type B strains. In order to elucidate if the expression of PilT could be correlated to the higher virulence of type A strains, we also constructed an in-frame deletion in the pilT gene.
We also attempted to make an in-frame deletion of the pilA gene, but in spite of several attempts we were unable to generate an unmarked deletion. It is possible that this is linked to the fact that there are two direct repeats flanking pilA and that this somehow affects the recombination in this region . We therefore chose a different strategy where we introduced a chloramphenicol resistance gene to allow for direct selection of the mutational event. In order to lower the risk of polar effects on the downstream pilE gene the resistance gene was inserted in the same orientation as the pilAE genes. We could also verify that the levels of pilE transcription were similar in the pilA mutant and wild-type strain, suggesting that there were no major polar effects on downstream genes. We have previously shown that pilV is transcribed from a promoter downstream of pilE gene in type strains  and also in this case pilV transcription levels were similar in the pilA and the wild-type strain. From this we conclude that none of the mutations generated any major polar effect on transcription of neighboring genes.
PilA expression in Tfp mutant strains
pilA, pilC and pilQ contribute to virulence of SCHU S4
Mice infection data.
Recently, we have provided evidence that PilA, one of the type IV pilins encoded by the human pathogenic subspecies of F. tularensis, is an important virulence determinant for type B strains . In addition, we have established that loss of the pilA gene is one of two major genetic events, responsible for the attenuation of the live vaccine strain, LVS [6, 24]. Even though we have been able to demonstrate PilA to be both surface located in F. tularensis and able to form functional Tfp in the heterologous system in Neisseria gonorrhoeae, we have still not managed to verify PilA to be an actual structural component of Tfp expressed by F. tularensis.
In this study, we present evidence that PilA and the Tfp assembly/secretion proteins, PilC and PilQ, are required for full virulence of the type A strain, SCHU S4, the most virulent subspecies of F. tularensis. In infections with individual mutants, we were unable to show that mutations of the putative Tfp genes resulted in a significant attenuation. However, when we conducted mixed infections, where the ability of the mutants to compete with the wild-type strain was assessed, it became more obvious that Tfp encoding genes may play a role in the virulence of SCHU S4. This is in line with our observation that pilA mutants in highly virulent clinical isolates of type B strains are less attenuated compared to type B strains with weaker virulence, like LVS [22, 24]. A general problem with the mouse infection model is that mice are highly susceptible to Francisella and do not discriminate between the virulence properties of different F. tularensis subspecies in the same way as the human infection. The emerging picture is that pilA mutants show less attenuation in the most pathogenic subspecies. Still, we believe that PilA, and potentially also Tfp, may play an important role in virulence. This theory is supported by the fact that LVS has lost pilA, and that this is one of the causes of its attenuation . When genomes of different subspecies are compared, one striking difference is that the pilT gene is a pseudogene in type B strains, due to a point mutation introducing a stop codon in the middle of the gene . Interestingly, in a study involving the attenuated type B strain LVS the pilT gene was demonstrated to be involved in pili assembly, adherence and virulence . Chakraborty with colleagues have suggested the possibility that the truncated PilT protein somehow has retained function in LVS . Their findings are somewhat surprising since in other Tfp expressing pathogens the PilT protein is only involved in pilus retraction and not in pilus assembly. The pilT mutant in SCHU S4 did not have any impact on the virulence in the subcutaneous mouse infection model. However, the fact that pilT is intact in most pathogenic type A strains suggests that PilT might, at least partly, contribute to the higher virulence of type A strains. As mentioned above, this infection model has limitations and may not reflect virulence in humans. Another possibility is that PilT may rather play a role in the environment and/or in transmission of tularemia than in the animal/human infection. With the genetic tools and the availability of specific mutants in the Tfp encoding gene clusters of SCHU S4, it will now be possible to address the role of the Tfp system in other infection models, for survival in the environment, and perchance for vector-borne transmission.
We have shown that pilA is required for full virulence of SCHU S4 in mice - a result in line with our earlier findings in type B strains. In addition, we have also demonstrated that the pilin assembly genes, pilC and pilQ, are needed for full virulence of SCHU S4. An unexpectedly finding is that PilT, even though it is functional only in type A strains, did not contribute to virulence in the mouse subcutaneous infection model.
Bacterial strains, plasmids, growth conditions, and DNA methods
Strains and plasmids used in this study
tularensis; SCHU S4
Human ulcer 1941, Ohio
FSC237; ΔpilA; deletion of codons 1-135
FSC237; ΔpilC (FTT1134); in frame deletion of codons 5-405
FSC237; ΔpilQ (FTT1156); in frame deletion of codons 13-593
FSC237; ΔpilT (FTT0088); in frame deletion of codons 7-336
F- mcrA Δ(mrr-hsdRMS-mcrBC), Φ80lacZ ΔM15 ΔlacX74 recA1 deoR araD139 (Δara-leu)7697 galU galK rpsL (Smr) endA1 nupG
recA, thi, pro, hsdR-M+,<RP4:2-Tc:Mu:Km:Tn7> TpR, SmR
TOPO-cloning vector. AmpR, KmR
Suicide plasmid. sacB; mobRP4; oriR6K; CmR
Suicide plasmid. groESL promoter, ori T, bla, sacB
432 bp fragment of pilA including a chlorampenicol resistance cassette cloned into pSMP22. CmR
2072 bp fragment of approximately 1 kb upstream and 1 kb downstream of pilC cloned in XbaI and SalI site of pDM4. CmR
2122 bp fragment of approximately 1 kb upstream and 1 kb downstream of pilQ cloned in XbaI and SalI site of pDM4. CmR
2133 bp fragment of approximately 1 kb upstream and 1 kb downstream of pilT cloned in XbaI and SalI site of pDM4. CmR
Primers used in this study
Primer sequence 5'-3'
Construction of deletion mutants of pilA, pilC, pilQ, and pilT in FSC237
Left and right flanking regions of pilA (FTT0890c, SCHU S4 nomenclature) were PCR amplified using the primer pairs pilA_LFF/pilA_LFR and pilA_RFF/pilA_RFR, and cloned into pGEMT-easy (Promega). The left flank was excised with EcoRI and PstI and the right flank was excised with BamHI and PstI. The fragments were ligated into an EcoRI/BamHI digested pBluescript KS+ vector (Stratagene), giving rise to pSMP47. A chloramphenicol resistance gene was PCR amplified from pDM4 with the primer pair CAM_PstIF/CAM_PstIR, digested with PstI, and cloned into pSMP47, generating pSMP48 containing the left and right flanks of pilA disrupted by a chloramphenicol cassette. The mutated allele of pilA was excised from pSMP48 with SphI and MluI, cloned into pSMP22, and the resulting plasmid pSMP50CAM (Table 2) was introduced into strain FSC237 by conjugal mating as previously described .
The deletion constructs for pilC, pilQ and pilT were obtained by overlapping PCR using the following primer pairs PilX_A, PilX_B, PilX_C and PilX_D (Table 3). For the deletion constructs of pilC and pilQ strain FSC237 was used as template and for the pilT deletion the strain FSC155 was used as a template. The sequence for the pilT construct is almost identical between FSC155 and FSC237 except for three substitutions upstream of the deletion in non-coding sequences, and eight substitutions in a downstream pseudogene. The PCR fragments were cloned into the suicide vector pDM4 and the resulting plasmids pAL12 (pilC), pAL16 (pilQ), and pAL18 (pilT) (Table 2) were introduced into strain FSC237 by conjugal mating as previously described . The in vitro growth rate of the different mutant strains were compared with the wild type strain by measuring OD at different time points, 0 h, 6 h and ON after dilution in Chamberlain medium.
RNA isolation and RT-PCR
Bacteria were grown for 18 h on plates, harvested and suspended in TRIzol reagent (Life Technologies). Total RNA was extracted and treated with RNase-free DNase I (Roche), phenol extracted, and precipitated by ethanol. An aliquot of the RNA (3 μg) was used to synthesize cDNA using random hexamers (final concentration 25 ng/μl) and Superscript III reverse transcriptase as described by the manufacturer (Life Technologies). In control experiments samples processed without addition of RT enzyme were used.
F. tularensis strains were grown for 16 h on BCGA before the bacteria were suspended in phosphate buffered saline (PBS) pH 7.4 to an OD540 = 1, which normally corresponds to approximately 2 × 109 bacteria/ml. The bacterial suspension was then diluted in PBS into two doses used for challenge, around 10 and 100 bacteria in a total volume of 100 μl. All bacterial infections were initiated by subcutaneous injections of 6-8 week old C57Black/6 female mice. The study was approved by the Local Ethical Committee on Laboratory Animals in Umeå, Sweden. For competitive index (CI) infections, the mice were infected with a 50:50 mixture of mutant and wild-type strains with around 50 bacteria of each strain. Mice were culled five days post-infection, and the spleens were homogenized in 1 ml of PBS and spread on BCGA. Individual colonies were analysed by PCR with primers specific for each mutation in order to examine the distribution of each strain. Spleens from at least three animals were collected for each pair of strains, and at least 200 colonies were analysed by PCR. The CI was calculated for each strain by dividing the ratio of mutant/wt after infection (determined with PCR) with the ratio of mutant/wt before infection (determined by viable count). Statistical analysis was performed with a GraphPad Prism computer software program using a paired Student's t-test (one-tailed) where P < 0.05 was regarded as significant.
Gel electrophoresis and Western blotting
Samples were boiled in the presence of SDS and Β-mercaptoethanol for 5 min and then separated on a 12% acrylamide gel by electrophoresis as described by Laemmli . Proteins were transferred to Immobilon-P Transfer Membranes using a Trans-Blot Semi-Dry transfer cell (BioRad). Membranes were blocked overnight in Tris-buffered saline (TBS) with 5% nonfat dry milk. Membranes were probed with rabbit polyclonal anti-PilA sera  and a horseradish peroxidase-conjugated anti rabbit antibody (Amersham Pharmacia Biotech) was used as secondary antibody and the filters were developed by using the ECL Kit (Amersham Pharmacia Biotech) according to the instructions from the manufacturer.
- Mörner T: The ecology of tularaemia. Rev Sci Tech. 1992, 11: 1123-1130.PubMedGoogle Scholar
- Tärnvik A: Nature of protective immunity to Francisella tularensis. Rev Infect Dis. 1989, 11: 440-451.View ArticlePubMedGoogle Scholar
- Petersen J, Schriefer M: Tularemia: emergence/re-emergence. Vet Res. 2005, 36: 455-467. 10.1051/vetres:2005006.View ArticlePubMedGoogle Scholar
- Whipp M, Davis J, Lum G, de Boer J, Zhou Y: Characterization of a novicida-like subspecies of Francisella tularensis isolated in Australia. J Med Microbiol. 2003, 52: 839-842. 10.1099/jmm.0.05245-0.View ArticlePubMedGoogle Scholar
- Larsson P, Oyston P, Chain P, Chu M, Duffield M: The complete genome sequence of Francisella tularensis, the causative agent of tularemia. Nat Genet. 2005, 37: 153-159. 10.1038/ng1499.View ArticlePubMedGoogle Scholar
- Rohmer L, Fong C, Abmayr S, Wasnick M, Larson Freeman T: Comparison of Francisella tularensis genomes reveals evolutionary events associated with the emergence of human pathogenic strains. Genome Biol. 2007, 8: R102-10.1186/gb-2007-8-6-r102.PubMed CentralView ArticlePubMedGoogle Scholar
- Golovliov I, Sjöstedt A, Mokrievich A, Pavlov V: A method for allelic replacement in Francisella tularensis. FEMS Microbiol Lett. 2003, 222: 273-280. 10.1016/S0378-1097(03)00313-6.View ArticlePubMedGoogle Scholar
- Fullner K, Mekalanos J: Genetic characterization of a new type IV-A pilus gene cluster found in both classical and El Tor biotypes of Vibrio cholerae. Infect Immun. 1999, 67: 1393-1404.PubMed CentralPubMedGoogle Scholar
- Mattick J, Whitchurch C, Alm R: The molecular genetics of type-4 fimbriae in Pseudomonas aeruginosa--a review. Gene. 1996, 179: 147-155. 10.1016/S0378-1119(96)00441-6.View ArticlePubMedGoogle Scholar
- Tønjum T, Koomey M: The pilus colonization factor of pathogenic neisserial species: organelle biogenesis and structure/function relationships--a review. Gene. 1997, 192: 155-163. 10.1016/S0378-1119(97)00018-8.View ArticlePubMedGoogle Scholar
- Strom M, Nunn D, Lory S: A single bifunctional enzyme, PilD, catalyzes cleavage and N-methylation of proteins belonging to the type IV pilin family. Proc Natl Acad Sci USA. 1993, 90: 2404-2408. 10.1073/pnas.90.6.2404.PubMed CentralView ArticlePubMedGoogle Scholar
- Helaine S, Dyer D, Nassif X, Pelicic V, Forest K: 3 D structure/function analysis of PilX reveals how minor pilins can modulate the virulence properties of type IV pili. Proc Natl Acad Sci USA. 2007, 104: 15888-15893. 10.1073/pnas.0707581104.PubMed CentralView ArticlePubMedGoogle Scholar
- Winther-Larsen H, Wolfgang M, Dunham S, van Putten J, Dorward D: A conserved set of pilin-like molecules controls type IV pilus dynamics and organelle-associated functions in Neisseria gonorrhoeae. Mol Microbiol. 2005, 56: 903-917. 10.1111/j.1365-2958.2005.04591.x.View ArticlePubMedGoogle Scholar
- Wolfgang M, van Putten J, Hayes S, Dorward D, Koomey M: Components and dynamics of fiber formation define a ubiquitous biogenesis pathway for bacterial pili. EMBO J. 2000, 19: 6408-6418. 10.1093/emboj/19.23.6408.PubMed CentralView ArticlePubMedGoogle Scholar
- Strom M, Lory S: Structure-function and biogenesis of the type IV pili. Annu Rev Microbiol. 1993, 47: 565-596. 10.1146/annurev.mi.47.100193.003025.View ArticlePubMedGoogle Scholar
- Whitchurch C, Hobbs M, Livingston S, Krishnapillai V, Mattick J: Characterisation of a Pseudomonas aeruginosa twitching motility gene and evidence for a specialised protein export system widespread in eubacteria. Gene. 1991, 101: 33-44. 10.1016/0378-1119(91)90221-V.View ArticlePubMedGoogle Scholar
- Skerker J, Berg H: Direct observation of extension and retraction of type IV pili. Proc Natl Acad Sci USA. 2001, 98: 6901-6904. 10.1073/pnas.121171698.PubMed CentralView ArticlePubMedGoogle Scholar
- Mattick J: Type IV pili and twitching motility. Annu Rev Microbiol. 2002, 56: 289-314. 10.1146/annurev.micro.56.012302.160938.View ArticlePubMedGoogle Scholar
- Chakraborty S, Monfett M, Maier T, Benach J, Frank D: Type IV pili in Francisella tularensis: roles of pilF and pilT in fiber assembly, host cell adherence, and virulence. Infect Immun. 2008, 76: 2852-2861. 10.1128/IAI.01726-07.PubMed CentralView ArticlePubMedGoogle Scholar
- Zogaj X, Chakraborty S, Liu J, Thanassi D, Klose K: Characterization of the Francisella tularensis subsp. novicida type IV pilus. Microbiology. 2008, 154: 2139-2150. 10.1099/mic.0.2008/018077-0.View ArticlePubMedGoogle Scholar
- Gil H, Benach J, Thanassi D: Presence of pili on the surface of Francisella tularensis. Infect Immun. 2004, 72: 3042-3047. 10.1128/IAI.72.5.3042-3047.2004.PubMed CentralView ArticlePubMedGoogle Scholar
- Forslund A, Kuoppa K, Svensson K, Salomonsson E, Johansson A: Direct repeat-mediated deletion of a type IV pilin gene results in major virulence attenuation of Francisella tularensis. Mol Microbiol. 2006, 59: 1818-1830. 10.1111/j.1365-2958.2006.05061.x.View ArticlePubMedGoogle Scholar
- Svensson K, Larsson P, Johansson D, Byström M, Forsman M: Evolution of subspecies of Francisella tularensis. J Bacteriol. 2005, 187: 3903-3908. 10.1128/JB.187.11.3903-3908.2005.PubMed CentralView ArticlePubMedGoogle Scholar
- Salomonsson E, Kuoppa K, Forslund A, Zingmark C, Golovliov I: Reintroduction of two deleted virulence loci restores full virulence to the live vaccine strain of Francisella tularensis. Infect Immun. 2009, 77: 3424-3431. 10.1128/IAI.00196-09.PubMed CentralView ArticlePubMedGoogle Scholar
- Hager A, Bolton D, Pelletier M, Brittnacher M, Gallagher L: Type IV pili-mediated secretion modulates Francisella virulence. Mol Microbiol. 2006, 62: 227-237. 10.1111/j.1365-2958.2006.05365.x.View ArticlePubMedGoogle Scholar
- Rohmer L, Brittnacher M, Svensson K, Buckley D, Haugen E: Potential source of Francisella tularensis live vaccine strain attenuation determined by genome comparison. Infect Immun. 2006, 74: 6895-6906. 10.1128/IAI.01006-06.PubMed CentralView ArticlePubMedGoogle Scholar
- Salomonsson E, Forsberg A, Roos N, Holz C, Maier B: Functional analyses of pilin-like proteins from Francisella tularensis: complementation of type IV pilus phenotypes in Neisseria gonorrhoeae. Microbiology. 2009, 155: 2546-2559. 10.1099/mic.0.028183-0.View ArticlePubMedGoogle Scholar
- Sambrook J, Fritsch EF, Maniatis T: Molecular cloning: a laboratory manual. 1989, Cold Spring Harbor: Cold Spring Harbor Laboratory Press, 654 s. med var. pag, 2Google Scholar
- Thomas R, Titball R, Oyston P, Griffin K, Waters E: The immunologically distinct O antigens from Francisella tularensis subspecies tularensis and Francisella novicida are both virulence determinants and protective antigens. Infect Immun. 2007, 75: 371-378. 10.1128/IAI.01241-06.PubMed CentralView ArticlePubMedGoogle Scholar
- Milton D, O'Toole R, Horstedt P, Wolf-Watz H: Flagellin A is essential for the virulence of Vibrio anguillarum. J Bacteriol. 1996, 178: 1310-1319.PubMed CentralPubMedGoogle Scholar
- Laemmli U: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970, 227: 680-685. 10.1038/227680a0.View ArticlePubMedGoogle Scholar
- Simon R, Priefer U, Pühler A: A Broad Host Range Mobilization System for In Vivo Genetic Engineering: Transposon Mutagenesis in Gram Negative Bacteria. Bio/Technology. 1983, 784-791. 10.1038/nbt1183-784.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.