Genotypic and phenotypic characterization of fusobacteria from Chinese and European patients with inflammatory periodontal diseases

ARTICLE IN PRESS

Systematic and Applied Microbiology 29 (2006) 120–130 www.elsevier.de/syapm

Genotypic and phenotypic characterization of fusobacteria from Chinese and European patients with inflammatory periodontal diseases$ Rudolf Gmu¨ra,, Mark A. Munsonb, William G. Wadeb a

Institute for Oral Biology, Center for Dental-, Oral Medicine and Maxillofacial Surgery, University of Zu¨rich, Plattenstrasse 11, CH-8032 Zu¨rich, Switzerland b Department of Microbiology, Dental Institute, King’s College London, Guy’s Campus, London SE1 9RT, UK Received 21 June 2005

Abstract Phylogenetic and antigenic studies were performed on 48 human oral Fusobacterium strains from Chinese patients with either necrotizing ulcerative gingivitis (NUG) or gingivitis and on 23 Fusobacterium nucleatum or Fusobacterium periodonticum strains from European periodontitis patients. Alignment of partial 16S rRNA gene sequences resulted in a phylogenetic tree that corresponded well with the current classification of oral fusobacteria into F. periodonticum and several subspecies of F. nucleatum, in spite of much minor genetic variability. F. periodonticum, F. nucleatum subsp. animalis and a previously undescribed phylogenetic cluster (C4), that may represent an additional F. nucleatum subspecies, constituted discrete clusters distinct from the remainder of F. nucleatum with high bootstrap values. Chinese and European strains differed markedly with regard to their respective classification patterns, suggesting a predominance of F. peridonticum and F. nucleatum susp. animalis over F. nucleatum subsp. nucleatum and F. nucleatum subsp. fusiforme/vincentii in samples from China. Antigenic typing enabled the association of many previously described serovars with distinct phylogenetic clusters and when applied directly to uncultured clinical samples confirmed the differential distribution of oral Fusobacterium taxa in Chinese and European samples. Bacteria from cluster C4 and F. nucleatum subsp. animalis were significantly more prevalent and accounted for higher cell numbers in NUG than in gingivitis samples, suggesting a possible association of these rarely observed taxa with NUG in Chinese patients. r 2005 Elsevier GmbH. All rights reserved. Keywords: Fusobacterium; 16s rDNA; Antigenic diversity; Phylogenetic diversity; Necrotizing ulcerative gingivitis; Gingivitis; Periodontitis

Abbreviations: CBA, Columbia blood agar; CBA-KV, Columbia blood agar with kanamycin and vancomycin; FAA-VEN, fastidious anaerobe agar with vancomycin; erythromycin and norfloxacin; FUM, fluid universal medium; IG, immunoglobulin; IF, immunofluorescence; MAbs, Monoclonal antibodies; NUG, necrotizing ulcerative gingivitis $ Note: Nucleotide sequence data reported are available in the DDBJ/EMBL/GenBank databases under the accession numbers AJ810270–AJ810282. Corresponding author E-mail address: [email protected] (R. Gmu¨r). 0723-2020/$ - see front matter r 2005 Elsevier GmbH. All rights reserved. doi:10.1016/j.syapm.2005.07.011

Introduction Fusobacterium nucleatum is the most frequently isolated species of the genus Fusobacterium in humans. Its presence has been reported in polymicrobial infections such as necrotizing ulcerative gingivitis (NUG), periodontitis, sinusitis, empyema, septicaemia, endocarditis, or periapical, cutaneous, subcutaneous, intra-abdominal, spinal and brain abscesses [1–3,32]. Its primary habitat

ARTICLE IN PRESS R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130

is thought to be the oral cavity, where it lives as a commensal preferentially colonizing the gingival sulcus and periodontal pockets but is capable of opportunistic pathogenicity in the mouth and other body sites [24]. Fusobacterium periodonticum [29], phenotypically barely distinguishable from F. nucleatum, is a much less studied species that colonizes the same complex oral biofilms. A range of genetic, phenotypic and antigenic properties of F. nucleatum have been investigated using planktonically grown cells from a small number of reference strains [2] and the genome of the type strain and one other strain have been sequenced [20,21]. However, to what extent the findings from such studies are representative of the species as a whole is far from clear since F. nucleatum is very heterogeneous. This is reflected by its current, repeatedly questioned, classification into five subspecies [6,10,12,13,27], by a proposal of an alternative classification scheme [26,27] and by the description of new species formerly being part of F. nucleatum [7,29]. Analyses of human clinical samples have revealed broad diversity with the presence of multiple genetically or antigenically distinguishable strains in the same ecological niche [12,19,25,31,33]. However, the biological and clinical significance of this diversity is still largely unknown. The present work had two principal aims: First we wanted to investigate the genetic relationship of strains from two previous studies [18,33]. Those of the latter study had been classified as F. nucleatum/F. periodonticum based on phenotypic traits, but serologically seemed to be of extensive heterogeneity, whereas for those of the former investigation little information was available. Our hypothesis was that this analysis should allow an association of serovars and phylotypes, a connection that had been elusive so far. A secondary aim was to determine the distribution of F. nucleatum sub-types in marginal plaque samples from Chinese patients with gingivitis or NUG.

Materials and methods Patients and plaque sampling Marginal supragingival plaque samples were collected with a sterile curette from buccal and/or oral surfaces of the most disease-affected regions of 19 patients with gingivitis and 21 patients with NUG [18]. Patients were recruited at dental clinics in Beidaihe, Chengede, Shijiazhuang and Xi’an (People’s Republic of China) where they signed informed consent for the collection of dental plaque for research purposes. Plaque from different sites of a subject was pooled into an Eppendorf centrifuge vial containing 1 ml of reduced transport fluid [23] with 10% glycerol. Samples were frozen at
70 1C

121

within 1 h from sampling, sent together on dry ice to Zu¨rich, and stored there at
75 1C until used.

Cultures and biochemical analyses Freshly defrosted samples were vortexed (45 s), sonicated for 5 s at 30 W (Sonifier B-12, Branson) and split into aliquots [18]. Serial dilutions of one aliquot were used to inoculate three media of increasing selectivity: (1) Columbia blood agar (CBA) (Oxoid) plates containing 5% (v/v) haemolysed human blood, (2) CBA plates with 0.1 mg kanamycin l
1 (Fluka) and 5 mg vancomycin l
1 (Eli Lilly) (CBA-KV), and (3) fastidious anaerobe agar plates with 4 mg vancomycin l
1, 1 mg erythromycin l
1 and 1 mg norfloxacin l
1 (both from Sigma-Aldrich) (FAA-VEN). Cultures were incubated for 96 h at 37 1C in an anaerobic chamber with 5% CO2, 10% H2, and 85% N2. Colonies were examined and enumerated using a stereomicroscope. Clonal strains from CBA-KV and FAA-VEN plates were coded [e.g. 6AK1 or 35AF4, whereby the initial number indicates a specific gingivitis (1–21) or NUG patient (31–52), AF or AK the selection medium (FAA-VEN or CBA-KV, respectively) and the last number the isolated colony], and frozen in liquid nitrogen. For preliminary speciation isolates were grown anaerobically for 24–48 h at 37 1C in fluid universal medium (FUM) [14] to an OD (550 nm) of 1. The pH of the cell free supernatant was determined from FUM cultures containing 0.3% glucose, the production of indole with FUM cultures lacking glucose [30]. Alkaline phosphatase activity was measured by monitoring the development of yellow colour following the incubation of a picked colony in 0.5 ml of substrate (30 mg p-nitrophenylphosphate dissolved in 1 ml of DMSO and supplemented with 10 ml of 0.1 M Tris-HCl, pH 8.5).

Antibodies, immunofluorescence Strains of fusiform cells were serologically characterized with monoclonal antibodies (MAbs) specific for F. nucleatum serovars [33], F. periodonticum serovars [33], Leptotrichia buccalis, or Capnocytophaga sp. MAbs 454CG1 (IgG3/IgM; double positive even after repetitive cloning) and 414LB1 (IgG2a) were generated as described [33] from splenocytes of two female Balb/c mice immunized intraperitonealy with approximately 108 pasteurized pooled bacteria from three Capnocytophaga gingivalis strains (ATCC 33624T, OMZ 574 and OMZ 686) or two L. buccalis strains (ATCC 14201T and OMZ 577), respectively. Specificity testing of 454CG1 by immunofluorescence (IF) demonstrated selective labelling of all Capnocytophaga strains tested (three strains of C. gingivalis, three of C. ochracea and five not further typed Capnocytophaga isolates). Analogously, 414LB1 bound to 13 of 14 L. buccalis strains tested

ARTICLE IN PRESS 122

R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130

and failed to react with reference strains from multiple other species. For IF of cultured strains, the cell suspensions were washed once in 0.9% NaCl, resuspended in 0.9% NaCl/0.02% NaN3/0.00025% cetyltrimethyl-ammonium bromide (all w/v), mounted on multi-well glass slides and stained by standard sandwich IF [15]. IF of plaque samples was carried out as described [16,17] and read by a single investigator (RG) using an Olympus BX60 epifluorescence microscope (Olympus Optical) equipped with phase-contrast, an HBO 103W/2 mercury photo optic lamp (Osram) and Olympus filter sets U-MNIBA (6-FAM), U-MA41007 (Cy3) and BXDFC5 (6-FAM/Cy3).

16S rDNA sequence analysis Phylogenetic analysis was performed on 42 strains from Chinese clinical samples, selected to represent the full phenotypic diversity observed with 84 isolates, that were tentatively identified as F. nucleatum or F. periodonticum (Table 1). The Chinese strains were augmented with 22 F. nucleatum/F. periodonticum strains of European origin and one strain presumably isolated in North America (FDC 364), most with a known antigenic profile [33]. The European strains originated from subgingival plaques collected in Germany (OMZ numbers 596–601), Sweden (OMZ 636, 643–645, 647), and Switzerland (OMZ 274, 439, 759, 766, 770–772, and 775–776). Reference strains OMZ 373 (FDC 364), OMZ

Table 1.

636 (ATCC 33693T), and OMZ 642 (ATCC 11326T) were received from the Forsyth Dental Center (Boston, USA), J. Carlsson (University of Umea( , Sweden), and the American Type Culture Collection (http:// www.lgcpromochem.com/atcc/), respectively. The 16S rDNA sequences of the Chinese strains and of eight complementary strains were determined by the London laboratory as described previously [9], those of the remaining 15 complementary strains by the Zurich laboratory using the following procedure: DNA was extracted from 5 ml FUM cultures with the GenElute Bacterial Genomic DNA Kit (Sigma-Aldrich) according to the manufacturer’s instructions. 16S rDNA was amplified using primers 27F and 1492R [22]. Reaction mixtures contained approximately 0.1 mg of genomic DNA, 1 mM of each primer, 0.5 mM dNTP’s, and 1 U of Vent DNA polymerase England (Bioconcept) in 1  Vent DNA polymerase reaction buffer. Amplifications were performed on a Biometra T-Gradient Thermocycler (Biolabo Scientific Instruments) with the following cycling conditions: initial denaturation at 95 1C for 5 min, 30 cycles of denaturation at 94 1C for 30 s, annealing at 54 1C for 30 s and extension at 72 1C for 2 min, followed by final extension at 72 1C for 5 min. The PCR products were directly sequenced (Microsynth) using primers 519R and 357F. A total of 885 bases (54 through 938) from the sequences determined in this study and from related sequences retrieved from GenBank were aligned by means of the Clustal X program [5] and further analysed with the PHYLIP

Results from biochemical and serological tests of 158 strains isolated from 42 Chinese gingivitis and NUG patients pH of FUM

APa

Indole

Binding of MAb

Presumptive identification

8

6.1–7.1 (1 strain o6)




+

X 1 antiFn/FpMAb

F. nucleatum or F. periodonticum

45

14

6.1–6.9 (1 strain 7.5)




+

None

F. nucleatum or F. periodonticum

5

17

4.6–5.7

+




454CG1

Capnocytophaga sp.

7/+




414LB1

L. buccalis

Disease group

Isolation medium

Gingivitis

NUG

FAA-VEN

7b

18

17

15

44

12

10

CBA-KV

(1 strain 6.9) 11

5

12

4

4.53–5.45 (1 strain 6.4)

0

4

4

0

6.4–6.5 (1 strain 5.6)

+

+

None




4.7–8.1

v




None




15

17

9

23

+, Positive or present;
, negative, absent, or unidentified; v, variable. a Alkaline phosphatase activity. b Number of isolates.

ARTICLE IN PRESS R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130

suite of programs [11]. Specifically, the DNADIST program was used to construct a distance matrix by use of the Jukes-Cantor algorithm, and the NEIGHBOR program was used to construct phylogenetic trees which were displayed using TREEVIEW [28].

Statistical analysis Prevalence scores of serovars in samples from the two test groups (gingivitis and NUG) were compared by Fisher’s exact test. Calculations were performed with StatView 5.01 (SAS Institute).

Results Phenotypic characterisation of isolates from Chinese NUG and gingivitis patients 158 strains were isolated from FAA-VEN and CBAKV plates and saved for further characterisation. A total of 84 of these strains, 22 from subjects with gingivitis and 62 from NUG patients, were presumptively identified as F. nucleatum or F. periodonticum on the basis of an absence of alkaline phosphatase activity, production of indole and failure to lower the pH below 6 in dense broth cultures containing glucose (Table 1), Serological characterisation with a large panel of MAbs used previously to characterize broad antigenic heterogeneity among F. nucleatum/F. periodonticum strains of European origin [33], resulted in only 25 strains (30%) being positive with at least one of the antibodies. A total of 38 saccharolytic isolates were identified as Capnocytophaga sp. or L. buccalis. A total of 36 other alkaline phosphatase positive or indole negative strains were not further investigated (Table 1).

16S rDNA sequence analysis Fig. 1 shows a phylogenetic tree that includes partial 16S rDNA sequences of 65 presumptive F. nucleatum or F. periodonticum strains and 18 strains or phylotypes retrieved from GenBank. The set of 65 test strains comprised 42 of the 84 isolates from Chinese NUG and gingivitis patients (Table 1) and 23 strains (OMZ numbers o900) from a previous study [33], The latter strains were selected to represent the broad spectrum of MAb-defined serovars. Tentatively, six principal clusters (labelled C1–C6) were identified, but generally bootstrap values were low. Only five branches had bootstrap values 490 (Fig. 1). However, with one exception, all reference strains for different Fusobacterium species or F. nucleatum subspecies fell into different clusters. The exception concerned cluster C2, which harboured in close proximity the type strains of F. nucleatum subsp.

123

fusiforme, F. nucleatum subsp. vincentii, and F. naviforme. Except for C4, all clusters seemed to correspond to one of the known F. nucleatum subspecies or to F. periodonticum. However, only F. nucleatum subsp. animalis (C3) and F. periodonticum (C6) were separated from the other F. nucleatum subsp. by deep branches. The small deeply branched cluster C4 harboured no reference strain. Within cluster C2 considerable diversity was evident, in particular a deep branch comprising seven isolates from this study and the uncultivated oral clone R002, but no reference strain. rDNA sequence variability occurred mainly within five regions of the gene (Table 2). Strains from clusters C5 and C6 had a 14-nucleotide deletion between bases 77 and 92 in common that was otherwise only seen with oral clones AJ050 (GenBank:AF287805), CZ006 (GenBank:AF287810) and OMZ 597, whereas strains from the other four clusters presented with unique cluster-specific sequences in that region. Similarly, cluster-specific motifs were found between nucleotides 183 and 201 (Table 2). Isolates from Chinese NUG and gingivitis patients were found dispersed in most clusters, but 84% of these strains fell into clusters C5 and C6, whereas 60% of the European strains grouped with clusters C1 and C2 (Table 3).

Association of F. nucleatum/F. periodonticum serovars with 16S rDNA sequence clusters Next, all strains printed in bold in Fig. 1 were tested for binding MAbs from a large panel of antibodies generated against F. nucleatum and F. periodonticum strains. The MAbs had been tested previously with more than 100 Fusobacterium strains and a similar number of strains from other taxa and none had shown reactivity with strains from taxa other than F. nucleatum/F. periodonticum (data not shown). Results with strains from Fig. 1 confirmed that MAbs 267FN1, 308FN1, 389FN1, 390FN1, 391FN1, and 393FN1 reacted selectively with isolates from the F. nucleatum subsp. fusiforme/vincentii cluster C2 (Fig. 2A) and thus appear to define subspecies-specific serovars. Further clusterspecific MAb-binding patterns were seen within the F. nucleatum subsp. nucleatum cluster C1 (Fig. 2B), the F. periodonticum cluster C6 (Fig. 2C), and cluster C4 (Fig. 2D). On the other hand, MAb 444FN1 recognized three of the four isolates that form a sub-branch of the F. nucleatum subsp. fusiforme/vincentii cluster C2 (Fig. 2A), but also labelled strain OMZ 982 from cluster C5. MAbs 311FN2.2, 375FN1, 395FN1 and 453FN1 bound all to strains from several clusters (data not shown). However, with the exception of 447FN1 (Fig. 2D), no MAb detected an epitope expressed by all members of an investigated cluster, subspecies, or species. 375FN1, for example, labelled strains 51AF1 (cluster

ARTICLE IN PRESS 124

R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130 F. necrophorum Fusobacterium russi ATCC 25533; M58681 Fusobacterium simiae ATCC 33568; M58685 Oral clone I035; AF287815 F. nucleatum subsp. animalis NCTC 12276T; X55404 96 OMZ 647 (JC-3:24) OMZ 990 (19AF2); AJ810279 36AF3 Oral clone CY024; AF287809 99 OMZ 775 OMZ 978 (34AF2); AJ810270 43AF1 78 99 F. nucleatum subsp. fusiforme NCTC 11326T; X55403 Oral clone AJ050; AF287805 F. nucleatum subsp. vincentii ATCC 49256; AJ006964 OMZ 776 OMZ 644 (JC-208); AJ810275 OMZ 770 OMZ 642 (NCTC 11326T) 65 Fusobacterium naviforme NCTC 11464T; AJ006965 55 OMZ 373 (FDC 364) OMZ 596 (KP-F2); AJ810276 OMZ 645 (JC1:27) 92 Oral clone R002; AF287806 OMZ 274; AJ810277 OMZ 759 40AF2 61 73 OMZ 772 OMZ 985 (43AF2); AJ810278 45AF2 OMZ 597 (KP-F5) 99 Oral clone CZ006; AF287810 OMZ 598 (KP-F8) 61 OMZ 643 (JC-2244) OMZ 771 F. nucleatum subsp. nucleatum ATCC 25586T; AJ133496 56 55 OMZ 439 F. nucleatum subsp. nucleatum ATCC 25586T; M58683 57 Fusobacterium canifelinum RMA 12708; AY162220 49AF2 OMZ 986 (43AK2); AJ810280 48AF2 31AK2 20AK1 10AF3 OMZ 601 (KP-F28) 57 OMZ 766; AJ810281 F. nucleatum subsp. polymorphum ATCC 10953T; AF287812 42AK3 39AF3 3AF2 OMZ 987 (47AF2); AJ810282 38AF2 33AK2 47AF3 12AF1 36AF2 32AK1 OMZ 984 (35AF2) 35AF1 OMZ 982 (52AK1); AJ810274 OMZ 983 (31AF4) 5AK1 6AK1 OMZ 989 (46AF3); AJ810273 48AF3 46AK2 75 F. periodonticum ATCC 33693T; X55405 OMZ 988 (52AF1) 51AF2 51AF1 OMZ 981 (13AF2); AJ810272 45AK2 21AF2 50AK1 42AF2 OMZ 979 (37AF2) F. periodonticum OMZ 636 (ATCC 33693T) OMZ 599 (KP-F10); AJ810271 OMZ 777 OMZ 980 (4AK1) Oral clone BS011; AF432130 OMZ 600 (KP-F12)

C3 C4

C2

C1

C5

85 97

0.01

C6

ARTICLE IN PRESS R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130

C6), 50AK1 (C6), OMZ 274 (C2), 48AF2 (C5) and OMZ 766 (C5) representing both F. nucleatum and F. periodonticum, but was negative with all other strains.

125

gingivitis samples contained 2.6 and NUG samples 4.4 serovars when investigated with 17 serovar-specific MAbs. The highest observed serovar diversity per sample was 6 for the gingivitis and 8 for the NUG group.

Serovars in NUG and gingivitis associated plaque Serovar distribution in Chinese supragingival plaques was investigated by IF using 23 MAbs. Fluorescence microscopy confirmed that all antibodies labelled only fusiform cells. Fig. 3 demonstrates, read column-bycolumn, that each MAb was positive with at least two of the 40 investigated NUG and gingivitis samples. The prevalence of bacteria identified by MAbs with specificity for clusters 1 or 2 (F. nucleatum subsp. nucleatum or F. nucleatum subsp. fusiforme/vincentii) was generally low and only few samples harboured positive bacteria in excess of 5  105 per sample (black fields in the first nine columns). Differences in prevalence between gingivitis and NUG patients were not significant. A much higher prevalence was found for fusobacteria identified by 372FN1, which binds to the F. nucleatum subsp. polymorphum type strain ATCC 10953, but again the difference between the disease groups was not significant. Higher prevalence and higher cell numbers of positive bacteria in the NUG than in the gingivitis group were seen with antibodies 422FN1/430FN1 and 447FN1/ 451FN1/452FN1. For both 422FN1- and 447FN1labelled bacteria the differences in prevalence between the two disease groups were significant (p-values of 0.001, and 0.045, respectively; Fisher’s exact test). We have tested by now more than 150 strains of human oral fusobacteria and so far 422FN1 and 430FN1 bound only to OMZ 647 from Cluster C3, whereas 447FN1, 451FN1 and 452FN1 reacted selectively with cluster C4 strains (data not shown). Antibodies 311FN2.2, 375FN1, 395FN1, 444FN1, and 453FN1 labelled fusobacteria of quite variable prevalence, but the cells were frequently present in the medium-to-high cell number range. All five MAbs are specific for epitopes shared by bacteria from different phylogenetic clusters or F. nucleatum subspecies [33]; data not shown. They were therefore designated panreactive MAbs in Fig. 3. Read line-by-line, Fig. 3 further demonstrates that all samples harboured multiple F. nucleatum and F. periodonticum serovars. On average,

Discussion The comparison of aligned partial 16S rDNA sequences from 65 human isolates and 18 clones retrieved from GenBank resulted in a phylogenetic tree characterized by considerable diversity, but an overall topology that corresponded well with the current classification of human oral fusobacteria. This is an interesting finding since this tree is based largely on newly determined sequences. All reference strains of different taxa fell into different branches of the tree, with the exception of the reference strains for F. nucleatum subsp. fusiforme and F. nucleatum subsp. vincentii, which grouped in cluster C2. Our results thus support the current classification scheme that is based on data from DNA–DNA hybridizations, SDS-PAGE protein profiles, and limited allozyme analysis [10,13,29] and which has been disputed repeatedly [26,27,33]. However, the bootstrap values weighting the robustness of the branching of the tree from Fig. 1 were generally low, especially with respect to the major branches separating the F. nucleatum subspecies nucleatum, fusiforme/vincentii, and polymorphum, and the recently described species F. canifelinum [7]. This underlines the heterogeneity of human oral fusobacteria and indicates that the tree topology should be regarded with caution. Cautious interpretation is also suggested by the surprising positioning in the tree of the uncultivated oral clones AJ050 and CZ006 and of isolate OMZ 597, which all share the otherwise cluster C5- and C6-specific deletion at positions 78–91 (Table 2). Seemingly strange was also the position of the reference strain for Fusobacterium naviforme, but the status of F. naviforme is in fact uncertain and requires re-evaluation as various reference strains from different sources apparently differ considerably and are not, or no longer, compatible with the original description of the species [8]. Our data suggest identity between F. naviforme NCTC 11464 and F. nucleatum subsp. fusiforme NCTC 11326, which is in

Fig. 1. Phylogenetic tree based on 16S rRNA gene sequence comparison over 885 aligned bases showing the relationship of strains from this study (bold face) and reference strains with sequences deposited in databanks (light face). Six clusters are marked (C1–C6) of which five harbour one or more reference strains for F. periodonticum (C6) or different subspecies of F. nucleatum (C1-3, C5). Cluster C4 did not harbour any speciated or sub-speciated reference strain and may represent a further F. nucleatum subspecies. The tree was constructed using the neighbour-joining method rooted for F. necrophorum following distance analysis of the aligned sequences. Numbers in italic represent bootstrap values for each branch based on data for 100 trees; only estimates 450 are shown. GenBank accession numbers are listed. The marker represents a 0.01% difference in nucleotide sequence.

ARTICLE IN PRESS 126

Table 2.

R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130

Alignment of selected 16S rDNA regions of F. nucleatum and F. periodonticum strains

Cluster F. nucleatum subsp. nucleatum (C1) F. nucleatum subsp. fusiforme/vincentii (C2) F. nucleatum subsp. animalis (C3) F. nucleatum subsp. (C4) F. nucleatum subsp. polymorphum (C5) F. periodonticum (C6)

Pos 77–92a

183–201

379–84

523–5

735–7

836–8

TTTTTTAACTTCGATT

ATTATAGGGCATCCTAGAA

CGAGAG

CGA

TGA

CAA

TTTTT-AACTTAGGTT

ATAATAGGGCATCCTATAA

CRAGAG

CGA

TGA

CAA

TTTTT-AACTTAGATT TCTTA-GACTTAGATT

ATTTTAAGGCATCTTAGAA ATTATACGGCATCGTATAA

CGAGAG CAAAAG

CGA CAA

TAA TAA

CAA CAA

T- - - - - - - - - - - - - - - - - - - T T- - - - - - - - - - - - - - - - - - - T

ATTTTAGGGCATCCTAARA ATTTTAGGGCATCCTAAGA

CAAGAG CAAAAG

CAA CAA

TRA TAA

CAA CTA

Isolates from the different clusters defined in Fig. 1 (C1–C6) display characteristic patterns at distinct positions printed in bold face. a Corresponding to positions in the E. coli 16S rDNA [4].

Table 3.

Distribution of the investigated strains in phylogenetic clusters C1–C6

Strains

Chinese NUG/gingivitis strains ðn ¼ 42Þ European periodontitis strains ðn ¼ 23Þ

Percentage of strains associated with cluster C1

C2

C3

C4

C5

C6

None

0 17

7 43

5 4

5 4

60 9

24 17

0 4

Fig. 2. Binding of MAbs by strains of clusters C2 (A), C1 (B), C6 (C), and C4 (D) defined in Fig. 1. Antibodies bound by the respective strain are listed to the right of the tree. Tested with 4150 F. nucleatum/F. periodonticum strains, all these MAbs displayed specificity for strains from a single cluster, except for 444FN1 that also labelled OMZ 982 of cluster C5. Strains/phylotypes printed in italics were not tested.

ARTICLE IN PRESS R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130

127

Monoclonal antibody Subject 305 428 424 389 393 391 267 308 390 372 302 422 430 383 385 447 451 452 311 375 395 444 453 G3 G4 G5 G6 G7 G8 G9 G10 G11 G12 G13 G14 G15 G16 G17 G18 G19 G20 G21 NUG31 NUG32 NUG33 NUG34 NUG35 NUG36 NUG37 NUG38 NUG39 NUG40 NUG42 NUG43 NUG44 NUG45 NUG46 NUG47 NUG48 NUG49 NUG50 NUG51 NUG52 F. nucleatum subsp. nucleatum

F. nucleatum subsp. fusiforme/vincentii

F. n.ss. polymor-phum

F. n. F. perio subsp. donticum animalis

F. nucleatum subsp.

pan-reactive MAbs

Fig. 3. Grid plot describing the prevalence and approximate cell number of MAb-labelled Fusobacterium populations in plaque samples from Chinese patients with NUG or gingivitis. The detected populations express taxon-specific epitopes of F. nucleatum subsp. nucleatum, F. nucleatum subsp. fusiforme/vincentii, F. nucleatum subsp. polymorphum, F. nucleatum subsp. animalis, F. periodonticum, or Fusobacterium sp. cluster C4, or bind panreactive MAbs, which detect epitopes shared by fusobacteria from several taxa. Key to the grey-level code of fields: white, no labelled bacteria detected; light grey, p5  104 positive bacteria ml
1 of sample; dark grey, 45  104 but p5  105 positive bacteria ml
1 of sample; black, 45  105 positive bacteria ml
1 of sample.

fact in agreement with the database of the National Collection of Type Culture (NCTC) that lists F. naviforme NCTC 11464 as a subclone of F. nucleatum subsp. fusiforme NCTC 11326. Three deep branches with high bootstrap values (497%) were of particular interest, namely those separating both the F. periodonticum (C6) and the F. nucleatum subsp. animalis (C3) clusters from the others and the one that gives weight to a small cluster (C4) formed by three strains that may correspond to a further subspecies of F. nucleatum. Notably, the three C4 strains expressed epitopes never

found on any other strain (see below). Once further isolates falling into this cluster are available, it will be interesting to assess its phylogenetic position more carefully using DNA–DNA hybridization analysis. A striking finding of the genetic analysis was the marked predominance of F. nucleatum subsp. polymorphum and F. periodonticum among the Chinese isolates and vice versa the predominance of F. nucleatum subsp. nucleatum and F. nucleatum subsp. fusiforme/vincentii among the isolates of European origin. It is a weakness of the present investigation that Chinese and European

ARTICLE IN PRESS 128

R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130

samples were collected from patients with different forms of periodontal disease, precluding the association of these differences with disease-specific characteristics, with distinct colonization patterns in subjects from different ethnicities that had lived separated for many millennia, or with cultural habits such as diet or oral hygiene. However, it should be noted that this study was not designed to address these questions as the Chinese samples derived from a study that had focused on the microbiology of NUG [18], a disease that nowadays is hardly ever found in Central European countries. The present finding of a predominance of F. periodonticum and F. nucleatum subsp. polymorphum strains among isolates from Chinese NUG and gingivitis patients is in agreement with earlier data obtained with these two patient groups [18]. In that study direct fluorescent in situ hybriditzation was used to enumerate fusobacteria in marginal plaque and on average less than half of all fusobacteria (labelled by the genus-specific Fusobacterium probe FUS664) were positive with a probe (Fnuc133) identifying specifically the F. nucleatum subspecies nucleatum, fusiforme/vincentii, and animalis, but neither F. periodonticum, nor most F. nucleatum subsp. polymorphum clones. Phenotypic characterization of the strains from this study included an analysis of their expression of epitopes detected by a panel of 23 antibodies. Some of these antibodies (e.g. 311FN2.2, 375FN2, 388FN1, 395FN1) are known to be pan-reactive [33] since they are bound by bacteria from different phylogenetic clusters. Others appear to define, however, F. periodonticum or F. nucleatum subspecies-specific (clusterspecific) serovars. With this study included, all MAbs described by Thurnheer et al. [33] have been tested on 4150 Fusobacterium strains. The data summarized in Fig. 2 demonstrate the highly restricted expression of many epitopes. Only the one recognized by 444FN1 was once detected on a strain from a ‘‘wrong’’ cluster. By combining antigenic and phylogenetic data, the results from this study corroborate our previous hypothesis that human oral fusobacteria express an almost strainspecific antigenic heterogeneity. The molecular identity of the antigens carrying these epitopes is not known. They have only been partially characterized and according to those data are polysaccharides, possibly of the O-antigenic side chain of the lipopolysaccharide [33]. Similarly, the function of such extreme diversity remains unknown. To the host’s immune system each serovar is an independent target, hence broad diversity could perhaps provide advantages in coping with the host response. Finally, we addressed the question to what extent these serovars, that now can be associated with distinct phylogenetic clusters, colonized in marginal plaque from Chinese patients with NUG or gingivitis. The results showed that all plaques harboured multiple antigeni-

cally distinct variants of F. nucleatum and F. periodonticum. Interestingly, bacteria that by this report’s combination of MAb-typing and rDNA sequencing became associated with the F. nucleatum subsp. animalis or the F. nucleatum cluster C4 were detected with significantly higher prevalence and with higher cell numbers per sample in plaques collected from NUG patients. Until now, these taxa were rarely detected (F. nucleatum subsp. animalis) or had not been described (C4). The current data provide evidence for marked differences among F. nucleatum subtypes in their association with two forms of inflammatory gingival disease. Whether this is related in any way to virulence properties or simply reflects ecological tropisms is not known and must be addressed by future studies. Another surprising observation was that several MAbs, whose specificity had appeared identical, revealed now differential epitope specificity. For example, antibodies 422FN1 and 430FN1 labelled in many samples (e.g. G17, G19, NUG37, etc.) similar numbers of fusobacteria, but in others (e.g. G10, G13, NUG31, etc.) only target cells reactive with the former antibody were detectable. Analogously, antibodies 447FN1, 451FN1, and 452FN1 were found to detect different epitopes that appear to be restricted to bacteria from cluster C4 (Figs. 2D and 3). Finally, the experiments showed that rather few samples harboured serologically detectable F. nucleatum subsp. nucleatum or F. nucleatum subsp. fusiforme/vincentii cells, whereas many more samples contained cells expressing antigens associated with F. nucleatum subsp. polymorhum or F. nucleatum subsp. animalis. Evidently, these results with uncultivated plaque samples agree very well with the data discussed above for isolated strains.

Acknowledgements The help of Yi Xue and Jan R. van der Ploeg in isolating and genetically characterizing strains from Chinese patients is gratefully acknowledged. We also thank Martin Gander, Yvonne Helweg and Verena Osterwalder for their excellent technical assistance.

References [1] K.W. Bennett, A. Eley, Fusobacteria: new taxonomy and related diseases, J. Med. Microbiol. 39 (1993) 246–254. [2] A.I. Bolstad, H.B. Jensen, V. Bakken, Taxonomy, biology, and periodontal aspects of Fusobacterium nucleatum, Clin. Microbiol. Rev. 9 (1996) 55–71. [3] I. Brook, E.H. Frazier, M.J. Gher, Microbiology of periapical abscesses and associated maxillary sinusitis, J. Periodontol. 67 (1996) 608–610. [4] J. Brosius, T.L. Dull, D.D. Sleeter, H.F. Noller, Gene organization and primary structure of a ribosomal RNA

ARTICLE IN PRESS R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

operon from Escherichia coli, J. Mol. Biol. 148 (1981) 107–127. R. Chenna, H. Sugawara, T. Koike, R. Lopez, T.J. Gibson, D.G. Higgins, J.D. Thompson, Multiple sequence alignment with the Clustal series of programs, Nucl. Acids Res. 31 (2003) 3497–3500. D.M. Citron, Update on the taxonomy and clinical aspects of the genus Fusobacterium, Clin. Infect. Dis. 35 (2002) S22–S27. G. Conrads, D.M. Citron, R. Mutters, S. Jang, E.J.C. Goldstein, Fusobacterium canifelium sp. nov., from the oral cavity of cats and dogs, Syst. Appl. Microbiol. 27 (2004) 4071–4413. G. Conrads, M.C. Claros, D.M. Citron, K.L. Tyrrell, V. Merriam, E.J.C. Goldstein, 16S–23S rDNA internal transcribed spacer sequences for analysis of the phylogenetic relationships among species of the genus Fusobacterium, Int. J. Syst. Evol. Microbiol. 52 (2002) 493–499. J. Downes, B. Olsvik, S.J. Hiom, D.A. Spratt, S.L. Cheeseman, I. Olsen, A.J. Weightman, W.G. Wade, Bulleidia extructa gen. nov., sp. nov., isolated from the oral cavity, Int. J. Syst. Evol. Microbiol. 50 (2000) 979–983. J.L. Dzink, M.T. Sheenan, S.S. Socransky, Proposal of three subspecies of Fusobacterium nucleatum Knorr 1922: Fusobacterium nucleatum subsp. nucleatum subsp. nov., comb. nov.; Fusobacterium nucleatum subsp. polymorphum subsp. nov., nom. rev., comb. nov.; and Fusobacterium nucleatum subsp. vincentii subsp. nov., nom. rev., comb. nov., Int. J. Syst. Bacteriol. 40 (1990) 74–78. J. Felsenstein, PHYLIP (Phylogeny Inference Package) version 3.5.c, Distributed by the author. Department of Genetics, University of Washington, Seattle, 1993. K.S. George, M.A. Reynolds, W.A.J. Falkler, Arbitrarily primed polymerase chain reaction fingerprinting and clonal analysis of oral Fusobacterium nucleatum isolates, Oral Microbiol. Immunol. 12 (1997) 219–226. S.E. Gharbia, H.N. Shah, Fusobacterium nucleatum subsp. fusiforme subsp. nov. and Fusobacterium nucleatum subsp. animalis subsp. nov. as additional subspecies within Fusobacterium nucleatum, Int. J. Syst. Bacteriol. 42 (1992) 296–298. R. Gmu¨r, B. Guggenheim, Antigenic heterogeneity of Bacteroides intermedius as recognized by monoclonal antibodies, Infect. Immun. 42 (1983) 459–470. R. Gmu¨r, B. Guggenheim, Interdental supragingival plaque – a natural habitat of Actinobacillus actinomycetemcomitans, Bacteroides forsythus, Campylobacter rectus, and Prevotella nigrescens, J. Dent. Res. 73 (1994) 1421–1428. R. Gmu¨r, J.R. Strub, B. Guggenheim, Prevalence of Bacteroides forsythus and Bacteroides gingivalis in subgingival plaque of prosthodontically treated patients on short recall, J. Periodont. Res. 24 (1989) 113–120. R. Gmu¨r, T. Thurnheer, Direct quantitative differentiation between Prevotella intermedia and Prevotella nigrescens in clinical samples, Microbiology 148 (2002) 1379–1387.

129

[18] R. Gmu¨r, C. Wyss, Y. Xue, T. Thurnheer, B. Guggenheim, Gingival crevice microbiota from Chinese patients with gingivitis or necrotizing ulcerative gingivitis, Eur. J. Oral Sci. 112 (2004) 33–41. [19] G. Haraldsson, W.P. Holbrook, E. Ko¨no¨nen, Clonal persistence of oral Fusobacterium nucleatum in infancy, J. Dent. Res. 83 (2004) 500–504. [20] V. Kapatral, I. Anderson, N. Ivanova, G. Reznik, T. Los, A. Lykidis, A. Bhattacharyya, A. Bartman, W. Gardner, G. Grechkin, L. Zhu, O. Vasieva, L. Chu, Y. Kogan, O. Chaga, E. Goltsman, A. Bernal, N. Larsen, M. D’Souza, T. Walunas, G. Pusch, R. Haselkorn, M. Fonstein, N. Kyrpides, R. Overbeek, Genome sequence and analysis of the oral bacterium Fusobacterium nucleatum strain ATCC 25586, J. Bacteriol. 184 (2002) 2005–2018. [21] V. Kapatral, N. Ivanova, I. Anderson, G. Reznik, A. Bhattacharyya, W.L. Gardner, N. Mikhailova, A. Lapidus, N. Larsen, M. D’Souza, T. Walunas, R. Haselkorn, R. Overbeek, N. Kyrpides, Genome analysis of F. nucleatum sub spp vincentii and its comparison with the genome of F. nucleatum ATCC 25586, Genome Res. 13 (2003) 1180–1189. [22] D.J. Lane, 16S/23S rRNA sequencing, In: E. Stackebrandt, M. Goodfellow (Eds.), Nucleic Acid Techniques in Bacterial Systematics, Wiley, Chichester, UK, 1991, pp. 115–175. [23] W.J. Loesche, R.N. Hockett, S.A. Syed, The predominant cultivable flora of tooth surface plaque removed from institutionalized subjects, Arch. Oral Biol. 17 (1972) 1311–1325. [24] W.E.C. Moore, L.H. Moore, R.R. Ranney, R.M. Smibert, J.A. Burmeister, H.A. Schenkein, The microflora of periodontal sites showing active destructive progression, J. Clin. Periodontol. 18 (1991) 729–739. [25] S.R. Moraes, J.F. Siqueira, I.N. Rocas, M.C.S. Ferreira, R.M.C.P. Domingues, Clonality of Fusobacterium nucleatum in root canal infections, Oral Microbiol. Immunol. 17 (2002) 394–396. [26] M.L. Morris, R.H. Andrews, A.H. Rogers, The use of allozyme electrophoresis to assess genetic heterogeneity among previously subspeciated isolates of Fusobacterium nucleatum, Oral Microbiol. Immunol. 11 (1996) 15–21. [27] M.L. Morris, R.H. Andrews, A.H. Rogers, Investigations of the taxonomy and systematics of Fusobacterium nucleatum using allozyme electrophoresis, Int. J. Syst. Bacteriol. 47 (1997) 103–110. [28] R.D.M. Page, TREEVIEW: an application to display phylogenetic tree on personal computers, Comput. Appl. Biosci. 12 (1996) 357–358. [29] J. Slots, T.V. Potts, P.A. Mashimo, Fusobacterium periodonticum, a new species from the human oral cavity, J. Dent. Res. 62 (1983) 960–963. [30] R.M. Smibert, N.R. Krieg, Phenotypic characterization, Methods for general and molecular biology In: P. Gerhardt, R.G.E. Murray, W.A. Wood, N.R. Krieg (Eds.), American Society of Microbiology, Washington, DC, 1994, pp. 607–654. [31] G. Suchett-Kaye, D. Decoret, O. Barsotti, Clonal analysis by ribotyping of Fusobacterium nucleatum isolates

ARTICLE IN PRESS 130

R. Gmu¨r et al. / Systematic and Applied Microbiology 29 (2006) 120–130

obtained from healthy young adults with optimal plaque control, J. Periodont. Res. 33 (1998) 179–186. [32] P.H. Summanen, D.A. Talan, C. Strong, M. McTeague, R. Bennion, J.J. Thompson, M.L. Vaisanen, G. Moran, M. Winer, S.M. Finegold, Bacteriology of skin and soft-tissue infections: comparison of infections in intravenous drug

users and individuals with no history of intravenous drug use, Clin. Infect. Dis. 20 (Suppl. 2) (1995) S279–S282. [33] T. Thurnheer, B. Guggenheim, B. Gruica, R. Gmu¨r, Infinite serovar and ribotype heterogeneity among oral Fusobacterium nucleatum strains?, Anaerobe 5 (1999) 79–92.