FEMS Microbiology Letters 179 (1999) 31^36

Molecular characterization of a £agellar (£a) operon in the oral

spirochete denticola ATCC 35405 Downloaded from https://academic.oup.com/femsle/article/179/1/31/605947 by guest on 27 September 2021

Lola V. Stamm *, Heather L. Bergen

Program in Infectious Diseases, Department of Epidemiology, School of Public Health, University of North Carolina, Chapel Hill, NC 27599-7400, USA

Received 5 May 1999; received in revised form 23 July 1999; accepted 23 July 1999

Abstract

A Treponema denticola 9.6-kb motility locus containing 11 genes was identified, sequenced and analyzed. The genes were designated tap1, flgD, flgE, orf4, motA, motB, fliL, fliM, fliY, orf10 and fliP. The order of these genes is identical to that of the corresponding region of the fla operon. Seven of the deduced Fla proteins share significant homology with both Escherichia coli and Bacillus subtilis proteins associated with flagellar structure and function. Reverse transcription-PCR analysis indicated that the T. denticola fla genes are transcribed as a single unit. A putative c28-like promoter, virtually identical to the T. pallidum fla promoter, was identified upstream of tap1. These results showed that the T. denticola and T. pallidum fla operons are highly conserved, supporting the proposed phylogenetic relatedness of these spirochetes. ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.

Keywords: Spirochete; Motility; ; Treponema denticola

1. Introduction wards the cell center [3]. The T. denticola PF are composed of a hook basal body complex and a £ag- Treponema denticola, a small anaerobic oral spiro- ellar ¢lament. The latter structure consists of three chete, is strongly associated with human periodontal core (FlaB) proteins and an outer sheath (FlaA) pro- disease [1]. Motility is an important virulence factor tein [4]. Little information is available about how the of this bacterium since it facilitates attachment to T. denticola PF are assembled. and invasion of periodontal tissues. Unlike the cell Genes encoding motility-associated proteins have surface £agella of Escherichia coli and Bacillus sub- been identi¢ed in pathogenic spirochetes and are tilis, spirochetal £agella are located within the peri- usually arranged in large operons. For example, the plasmic space [2]. In T. denticola, 2^3 periplasmic £a operon of Treponema pallidum ( agent) £agella (PF) originate at each end of the cell and [5^7] and the £gB operon of burgdorferi entwine the protoplasmic cylinder, extending to- ( agent) [8,9] contain 17 and 26 genes, respectively. Additionally, a £gB operon has been identi¢ed in T. denticola that contains at least eight * Corresponding author. Tel.: +1 (919) 966 3882; genes [10]. The 3P end of this operon has not been Fax: +1 (919) 966 2089; E-mail: [email protected] delineated. Because of our interest in spirochete mo-

0378-1097 / 99 / $20.00 ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S0378-1097(99)00387-0

FEMSLE 8973 9-9-99 32 L.V. Stamm, H.L. Bergen / FEMS Microbiology Letters 179 (1999) 31^36 tility, we sought to identify and characterize addi- denticola £gE [12] to amplify with PCR (Expand tional motility genes of T. denticola. We report Long Template PCR System kit, Boehringer- here the genetic and transcriptional analysis of the Mannheim, Indianapolis, IN, USA) an V7-kb re- T. denticola £a operon. gion of T. denticola genomic DNA (Wizard Genomic DNA Isolation kit, Promega). The PCR product was gel-puri¢ed, concentrated and used as template for 2. Materials and methods automated DNA sequencing using the Taq Dye- Deoxy Terminator Cycle Sequencing kit (PE Applied 2.1. Bacterial strains, plasmids, bacteriophages and Biosystems, Foster City, CA, USA) at the University growth conditions of North Carolina Automated DNA Sequencing Fa- Downloaded from https://academic.oup.com/femsle/article/179/1/31/605947 by guest on 27 September 2021 cility (Chapel Hill, NC, USA) on a model 377 DNA T. denticola ATCC 35405 (American Type Culture sequencer (PE Applied Biosystems). Both DNA Collection, Manassas, VA, USA) was grown anaer- strands were sequenced for accuracy. The DNA se- obically (BBL GasPak Plus, Becton Dickinson, quence 5P of the £gE gene was determined by screen- Cockeysville, MD, USA) at 37³C in BBL Mycoplas- ing the V ZapII T. denticola genomic DNA library ma broth (Becton Dickinson) supplemented with with £uorescein-labelled oligonucleotides. Probe se- 10% heat-inactivated rabbit serum (Gibco BRL, Be- quences were initially based on the partial sequence thesda, MD, USA) and 10 Wgml31 cocarboxylase. E. of £gE (£gEF1) and subsequently based on the spe- coli DH5K (Gibco BRL) and E. coli SOLR (Strata- ci¢c sequence of treponemal inserts from the V ZapII gene, La Jolla, CA, USA) were used as host strains genomic DNA library (tap1P, 5P-CTCCATTTCT- for preparation of plasmid DNA (Wizard Plus Mini- TACAGGTTCC-3P). prep DNA Puri¢cation System, Promega, Madison, WI, USA) for DNA sequence determinations. E. coli 2.3. Sequence analysis strains were grown at 37³C in Luria-Bertani (LB) broth and on LB agar containing 100 Wgml31 am- DNA and amino acid sequence analyses were per- picillin. E. coli XL1-Blue MRFP (Stratagene) was formed using the AssemblyLIGN Program (Interna- used for V ZapII infections according to the manu- tional Biotechnologies, New Haven, CT, USA), facturer's protocol. E. coli SOLR was used for ex- MacVector 4.5.1 (International Biotechnologies) cision of pBluescript SK3 (Stratagene) phagemid and the Genetics Computer Group (GCG) software from V ZapII. package version 9.1 (University of Wisconsin Bio- technology Center, Madison, WI, USA). The amino 2.2. Identi¢cation of the T. denticola £a operon acid sequences were used to search protein databases at the National Center for Biotechnology Informa- To identify £a genes, a V ZapII T. denticola ATCC tion (National Library of Medicine, Washington, 35405 genomic DNA library (provided by B. DC) using the basic local alignment search tool McBride, [11]) was probed with a £uorescein-labelled (BLAST) algorithm [13]. (Direct labeling kit, Amersham Life Sciences, Arling- ton Heights, IL, USA) 0.74-kb fragment of plasmid 2.4. Reverse transcription-PCR (RT-PCR) analysis p4B8P4 containing the T. pallidum £iP gene [5]. Tre- ponemal DNA inserts from positive recombinant V Total RNA was isolated from T. denticola ATCC phage were excised and re-circularized in pBluescript 35405 cells using TRI Reagent (Molecular Research SK3 according to the manufacturer's protocol. The Center, Cincinnati, OH, USA) according to the man- partial sequence of the T. denticola £iP gene ob- ufacturer's protocol. The RNA was treated with tained from the plasmids was used to design an oli- DNAse I (Gibco BRL) as recommended by the man- gonucleotide primer (£iPR1, 5P-GAGTCCCCAG- ufacturer. RT-PCRs (RT-PCR Access System kit, CCGTCAAC-3P). This primer was used with a Promega) were performed using oligonucleotide second primer (£gEF1, 5P-TAAACTATGCTTGTA- primer pairs (Pathology Oligonucleotide Facility, ACCTTG-3P) based on the partial sequence of T. University of North Carolina) that span the inter-

FEMSLE 8973 9-9-99 L.V. Stamm, H.L. Bergen / FEMS Microbiology Letters 179 (1999) 31^36 33 genic regions of the T. denticola £a genes. Additional lar mass for each of the deduced proteins encoded by RT-PCRs were performed using oligonucleotide pri- the ORFs are shown in Table 1. Putative ribosome mer pairs that span the region 5P of the promoter to binding sites precede each of the ORFs (GenBank within tap1 or 3P of the promoter to within tap1. For accession number AF122909). Intergenic spacing each oligonucleotide primer pair, PCRs were per- among the ORFs is very compact, ranging from formed with an RT reaction that lacked reverse tran- one to 42 nucleotides. Additionally, ORF8 overlaps scriptase, an RT reaction that contained reverse ORF9 by eight nucleotides and ORF10 overlaps transcriptase and with T. denticola genomic DNA ORF11 by four nucleotides. The G+C content of as a control. Optimization of RT reactions was per- the 9.6-kb region is 41%. The reported G+C content formed when necessary according to the manufactur- of the T. denticola genome is 37^38% [3]. Downloaded from https://academic.oup.com/femsle/article/179/1/31/605947 by guest on 27 September 2021 er's instructions (Promega Technical Bulletin #220, Based on signi¢cant nucleotide sequence homol- Promega). The resulting DNA products were electro- ogy of ORF1^11 with genes in the T. pallidum £a phoresed on a 2% agarose gel, stained with ethidium operon [5] and the partially characterized T. phage- bromide and photographed. denis £a operon [6], we have designated ORF1^9 and ORF11 as tap1, £gD, £gE, orf4, motA, motB, £iL, £iM, £iY and £iP, respectively. Although ORF10 3. Results and discussion corresponds to orf3 of the T. pallidum £a operon, we have retained the orf10 designation in accordance 3.1. Identi¢cation and characterization of the with the T. denticola ORF order. Identi¢cation of T. denticola £a genes tap1, £gD and £gE genes in another T. denticola strain (ATCC 33520) has recently been reported by Using a combination of PCR ampli¢cation of ge- Limberger et al. [14]. nomic DNA and screening of a genomic DNA li- A gene arrangement similar to that of the Trepo- brary, we obtained a 9.6-kb region of T. denticola nema £a operons is present in the B. burgdorferi £gB ATCC 35405 DNA. Analysis of the nucleotide se- operon [8]. Transcription of the £gB operon is initi- quence of this region indicated the presence of ated by a c70-like promoter several genes upstream 11 complete open reading frames (ORFs) with the of the tap1 homolog, £bC. Interestingly, Limberger same transcriptional orientation (Fig. 1). ORF1^11 et al. [6,14] identi¢ed by primer extension a c28-like are 1374, 510, 1392, 198, 780, 720, 546, 1035, 1182, promoter sequence (Pfla) preceding T. phagedenis and 660 and 816 nucleotides long, respectively (GenBank T. denticola (ATCC 33520) tap1. A similar promoter accession number AF122909). The size and molecu- sequence was also noted preceding T. pallidum tap1.

Fig. 1. Restriction and transcription maps of the T. denticola ATCC 35405 £a operon. (A) Partial restriction map with selected enzymes. (B) Gene map. The arrows indicate the location and direction of transcription of the £a genes.

FEMSLE 8973 9-9-99 34 L.V. Stamm, H.L. Bergen / FEMS Microbiology Letters 179 (1999) 31^36

Fig. 2. Comparison of the putative T. denticola ATCC 35405 c28-like £a promoter with the T. pallidum and T. phagedenis £a operon pro- 28 moters (Pfla) and the consensus c promoters of B. subtilis and E. coli/S. typhimurium. Downloaded from https://academic.oup.com/femsle/article/179/1/31/605947 by guest on 27 September 2021 We have identi¢ed a sequence 179 nucleotides up- that is homologous to the T. pallidum £a operon. stream of T. denticola tap1 whose 310 and 335 el- Conservation of the treponemal £a operons supports ements and spacing (16 nucleotides) are identical to the proposed phylogenetic relatedness of these spiro- those of T. denticola ATCC 33520 Pfla and virtually chetes [15]. 70 identical to those of T. pallidum Pfla (Fig. 2). A c - like promoter was not present in this region. 3.3. Amino acid sequence homologies

3.2. Transcriptional analysis The deduced amino acid sequences of the T. den- ticola Tap1-FliP proteins were used to conduct The presence of a putative c28-like promoter up- BLAST searches [13]. Table 1 shows the percentage stream of T. denticola tap1, the transcriptional ori- identity/similarity to selected homologs in T. pallid- entation of tap1^£iP and the absence of a transcrip- um, T. phagedenis, B. burgdorferi, E. coli and B. sub- tional terminator immediately following £iP strongly tilis. The highest homologies were observed with the suggested that these genes comprise part of an oper- corresponding spirochetal Fla proteins, demonstrat- on. Although the T. pallidum £a operon contains six ing their overall conservation. However, of all the additional genes downstream of £iP, this region has T. denticola Fla proteins, Tap1, which is unique to not been de¢ned yet in T. denticola. To identify the spirochetes, had the lowest percentage identity when nucleotide region containing the putative £a pro- compared to its spirochetal homologs. A multiple moter, RT-PCRs were performed with oligonucleo- tide primer pairs that span the regions 5P of the c28-like promoter to within tap1 or 3P of the c28-like promoter (after the putative transcriptional start site) to within tap1 (Fig. 3, sets 1 and 2, respec- tively). Only the latter reaction produced a DNA product of the expected size (Fig. 3, set 2, lane (+)), suggesting that transcription is initiated down- stream of the c28-like promoter. RT-PCRs were also performed to analyze transcription of the T. dentico- la tap1^£iP genes (Fig. 3, sets 3^11). Each of the oligonucleotide primer pairs used to span the inter- genic regions downstream of tap1 generated a DNA Fig. 3. RT-PCR analysis of the T. denticola ATCC 35405 £a op- product (Fig. 3, sets 3^11, lanes (+)), equivalent in eron. RT-PCRs were performed with oligonucleotide primer pairs size to the PCR product obtained with genomic spanning the following regions: (1) 5P £a promoter^tap1, (2) 3P DNA (Fig. 3, sets 3^11, lanes (C)). These products £a promoter^tap1, (3) tap1^£gD, (4) £gD^£gE, (5) £gE^motA, were not produced in the absence of RT (Fig. 3, sets (6) motA^motB, (7) motB^£iL, (8) £iL^£iM, (9) £iM^£iY, (10) 3^11, lanes (3)). Our results indicated that tap1^£iP £iY^orf10, (11) orf10^£iP. For each primer pair, PCRs were per- formed with a RT reaction that lacked reverse transcriptase (3), are part of a single mRNA transcript, suggesting a RT reaction that contained reverse transcriptase (+) or with that these genes are part of a large motility operon T. denticola genomic DNA as a control (C).

FEMSLE 8973 9-9-99 L.V. Stamm, H.L. Bergen / FEMS Microbiology Letters 179 (1999) 31^36 35

Table 1 Comparison of T. denticola Fla proteins Protein Function Size kDa % Identity/similarity to homologsa (amino acids) T. pallidum T. phagedenis B. burgdorferi E. coli B. subtilis Tap1 Unknown 457 51.0 21.9/32.1 32.8/41.2 21.0/30.5 ^b ^b FlgD Hook assembly 169 18.7 48.0/54.0 59.8/64.2 37.9/47.6 24.3/29.3 28.2/36.3 FlgE Hook 463 49.6 77.1/83.8 82.7/87.9 56.2/65.8 35.8/42.8 30.9/41.2 Orf4 Unknown 65 7.7 60.0/70.8 78.5/84.6 43.1/63.1 ^b ^b MotA Motor 259 28.3 69.5/81.5 72.6/80.3 43.6/54.4 26.9/36.2 34.4/42.5

MotB Motor 239 25.8 69.3/76.5 71.0/77.3 41.0/52.3 28.3/35.6 28.8/39.0 Downloaded from https://academic.oup.com/femsle/article/179/1/31/605947 by guest on 27 September 2021 FliL Unknown 181 20.6 52.8/68.5 57.1/71.2 34.1/46.2 ^b 16.1/30.7 FliM Switch 344 38.5 75.0/83.4 91.8/95.5c 57.6/67.4 28.2/37.9 40.8/54.4 FliY Switch 393 41.7 58.4/64.2 ^d 53.6/63.4 38.1/51.5 29.5/39.1 Orf10 Unknown 219 23.7 26.0/37.9 ^d 21.0/31.2 ^b 19.8/29.7 FliP Export 271 30.9 67.2/77.6 ^d 52.2/62.9 40.5/52.1 48.2/63.6 aBased on the GAP program from the GCG software package (BLOSUM60 matrix). bNo protein with signi¢cant overall amino acid sequence homology identi¢ed. cBased on the partial coding sequence. dData not available. sequence alignment of T. denticola Tap1 with T. pal- the role of the corresponding Fla proteins in trepo- lidum and T. phagedenis Tap1 revealed a well-con- nemal motility. served 90-amino acid region near the C-terminus of the protein, with minimal identity in the remainder of the protein (data not shown). This observation Acknowledgements suggested that the conserved C-terminus is important for the structure and/or function of Tap1. We thank B. McBride and P. Hannam for the V Based on comparative amino acid sequence anal- ZapII T. denticola genomic DNA library and S. yses, the T. denticola FlgD, FlgE, MotA, MotB, Greene for helpful suggestions. This research was FliM, FliY and FliP proteins presumably function supported by National Institutes of Health Grant in a manner analogous to their E. coli/B. subtilis U19 AI31496. homologs (Table 1). The functions of Orf4, FliL and Orf10 are currently unknown, although the lat- ter two proteins have limited homology to the B. References subtilis motility proteins FliL and FliZ, respectively. Until recently, analysis of the protein function in T. [1] Simonson, L.G., Goodman, C.H., Bial, J.J. and Morton, H.E. denticola has been hindered by the lack of genetic (1988) Quantitative relationship of Treponema denticola systems. However, Li et al. [12], using a gene inacti- to severity of periodontal disease. Infect. Immun. 56, 726^ vation method, demonstrated that T. denticola £gE 728. mutants (FlgE3) lack PF and are non-motile. Using [2] Holt, S.C. (1978) Anatomy and chemistry of spirochetes. Mi- crobiol. Rev. 42, 114^160. a similar approach, Limberger et al. [14] showed that [3] Smibert, R.M. (1984) Genus III: Treponema Schaudinn 1905, Tap1-de¢cient mutants are non-motile. They pro- 1728AL. In: Bergey's Manual of Systematic Bacteriology, Vol. posed that Tap1 is involved in monitoring the £ag- 1 (Krieg, N.R. and Holt, J.G., Eds.), pp. 49^57. Williams and ellar hook length, a function performed by FliK in Wilkins, Baltimore, MD. enteric . Our studies, which have provided [4] Ruby, J.D., Li, H., Kuramitsu, H., Norris, S.J., Goldstein, S.F., Buttle, K.F. and Charon, N.W. (1997) Relationship of information on eight newly identi¢ed T. denticola Treponema denticola periplasmic £agella to irregular cell mor- £a genes (orf4^£iP), should facilitate further use of phology. J. Bacteriol. 179, 1628^1635. gene inactivation to construct mutants for evaluating [5] Hardham, J.M., Frye, J.G. and Stamm, L.V. (1995) Identi¢-

FEMSLE 8973 9-9-99 36 L.V. Stamm, H.L. Bergen / FEMS Microbiology Letters 179 (1999) 31^36

cation and sequences of the Treponema pallidum £iM', £iY, K., Hatch, B., Smith, H.O. and Venter, J.C. (1997) Genomic £iP, £iQ, £iR and £hB' genes. Gene 166, 57^64. sequence of a Lyme disease spirochete, . [6] Limberger, R.J., Slivienski, L.L., El-Afandi, M.C.T. and Dan- Nature 390, 580^586. tuono, L.A. (1996) Organization, transcription and expression [10] Heinzerling, H.F., Olivares, M. and Burne, R.A. (1997) Ge- of the 5P region of the £a operon of Treponema phagedenis and netic and transcriptional analysis of £gB £agellar operon con- Treponema pallidum. J. Bacteriol. 178, 4628^4634. stituents in the oral spirochete Treponema denticola and their [7] Fraser, C.M., Norris, S.J., Weinstock, G.M., White, O., Sut- heterologous expression in enteric bacteria. Infect. Immun. 65, ton, G.G., Dodson, R., Gwinn, M., Hickey, E.K., Clayton, 2041^2051. R., Ketchum, K.A., Sodergren, E., Hardham, J.M., McLeod, [11] Haapasalo, M., Muller, K.H., Ultto, V.J., Leung, W.K. and M.P., Salzberg, S., Peterson, J., Khalak, H., Richardson, D., McBride, B.C. (1992) Characterization, cloning and binding Howell, J.K., Chidambaram, M., Utterback, T., McDonald, properties of the major 53-kilodalton Treponema denticola sur- L., Artiach, P., Bowman, C., Cotton, M.D., Fujii, C., Gar- face antigen. Infect. Immun. 60, 2058^2065. Downloaded from https://academic.oup.com/femsle/article/179/1/31/605947 by guest on 27 September 2021 land, S., Hatch, B., Horst, K., Roberts, K., Sandusky, M., [12] Li, H., Ruby, J., Charon, N. and Kuramitsu, H. (1996) Gene Weidman, J., Smith, H.O. and Venter, J.C. (1998) Complete inactivation in the oral spirochete Treponema denticola: con- genome sequence of Treponema pallidum, the syphilis spiro- struction of a £gE mutant. J. Bacteriol. 178, 3664^3667. chete. Science 281, 375^388. [13] Altschul, S.F., Madden, T., Scha¡er, A.A., Zhang, Z., Miller, [8] Ge, Y., Old, I.G., Saint Girons, I. and Charon, N.W. (1997) W. and Lipman, D.J. (1997) Gapped BLAST and PSI- Molecular characterization of a large Borrelia burgdorferi mo- BLAST: a new generation of protein database search pro- tility operon which is initiated by a consensus c70 promoter. grams. Nucleic Acids Res. 25, 3389^3402. J. Bacteriol. 179, 2289^2299. [14] Limberger, R.J., Slivienski, L.L., Izard, J. and Samsono¡, [9] Fraser, C.M., Casjens, S., Huang, W.M., Sutton, G.G., Clay- W.A. (1999) Insertional inactivation of Treponema denticola ton, R., Lathigra, R., White, O., Ketchum, K.A., Dobson, R., tap1 results in a nonmotile mutant with elongated £agellar Hickey, E.K., Gwinn, M., Dougherty, B., Tomb, J.-R., hooks. J. Bacteriol. 181, 3743^3750. Fleischmann, R.D., Richardson, D., Peterson, J., Kerlavage, [15] Paster, B.J., Dewhirst, F.E., Weisburg, W.G., Tordo¡, L.A., A.R., Quackenbush, J., Salzberg, S., Hanson, M., Vugt, R.V., Fraser, G.J., Hespell, R.B., Stanton, T.B., Zablen, L., Man- Palmer, N., Adams, M.D., Gocayne, J., Weidman, J., Utter- delco, L. and Woese, C.R. (1991) Phylogenetic analysis of back, T., Watthey, L., McDonald, L., Artiach, P., Bowman, spirochetes. J. Bacteriol. 173, 6101^6109. C., Garland, S., Fujii, C., Cotton, M.D., Horst, K., Roberts,

FEMSLE 8973 9-9-99