Genetic Relatedness of Bordetella Species as Determined by Macrorestriction Digests Resolved by Pulsed-Field Gel Electrophoresis

Transcription

1 INTERNATIONAL JOURNAL OF SYSTEMATI BATERIOLOGY, Oct. 99, p /9/ $02.00/0 opyright 0 99, International Union of Microbiological Societies Vol., No. Genetic Relateness of Boretella Species as Determine by Macrorestriction Digests Resolve by PulseFiel Gel Electrophoresis M. N. KHATTAKAND R.. MATTHEWS* Pertussis Reference Laboratory an Department of Meical Microbiology, Manchester University Meical School, Stopfor Builing, wor Roa, Manchester M 9PT, Unite Kingom The genetic relateness of the three meically important Boreteh species was examine by macrorestriction igestion of DNA with the rarely cutting enzyme XbaI an resolution by pulsefiel gel electrophoresis. Our ata showe that BoreteUa pertussis, Boretella parapertussis, an BoreteUa bronchiseptica prouce speciesspecific macrorestriction profiles an that there was some variation between ifferent isolates of the same species, onserve bans at 0 an 55 occurre with B. pertussis (0 isolates teste), but the nine variable bans between 200 an 2 istinguishe 2 types. The 0 isolates of B. parapertussis teste prouce up to bans at 8 to 75, of which were variable, giving three types The eight B. bronchiseptica isolates examine prouce up to 6 bans at 8 to 9, of which were variable, giving three types. The results of this work are compare with the results of previous DNADNA hybriization an multilocus enzyme electrophoresis stuies which suggeste that these three species are closely relate an shoul be consiere members of the same species. The following three meically important Boretella species have been escribe: Boretella pertussis, which causes whooping cough in humans; Boretella parapertussis, the agent which can be responsible for a mil whooping coughlike isorer in humans; an Boretella bronchiseptica, an animal respiratory pathogen which is very rarely isolate from humans. These are serologically relate bacteria with similar morphologies, sizes, an staining reactions (). They can be ifferentiate on the basis of their cultural characteristics, nutritional requirements, biochemical activities, an ranges of surface antigens (2). For in vitro culture from nasopharyngeal secretions, B. pertussis is grown on a meium such as cephelexin charcoal bloo agar, on which it prouces small, shiny colonies. B. pertussis is a slowly growing, nonmotile organism that prouces oxiase an catalase. The urease test is negative. B. parapertussis grows well on bloo agar an nutrient agar an is nonmotile. Urease is prouce after 2 h. atalase is prouce, while oxiase is not prouce. B. bronchiseptica grows as glistening betahemolytic colonies on bloo agar. This organism is motile, urease is prouce after h, an oxiase an catalase tests are positive. All three species ferment glucose an lactose with aci prouction (2). DNADNA hybriization (7) an multilocus enzyme electrophoresis stuies (0) have shown that these three species are closely relate. It has been suggeste that they shoul be consiere members of a single species. Restriction fragment length polymorphism, generate by frequently cutting enzymes, has been escribe as a highly iscriminatory system for etermining phylogenetic relationships among isolates of the same species an relate species (). This technique faile to iscriminate among isolates of B. pertussis, B. parapertussk, an B. bronchiseptica when DNAs were cut with EcoRI; the three organisms were inistinguishable (ata not shown). Pulsefiel gel electrophoresis (PFGE) provies an alternative tool for analyzing relationships among strains an * orresponing author. species by facilitating the stuy of large fragments of DNA (, 8). This technique was use to examine B. parapertussis an B. bronchiseptica an their relationship to B. pertzusis. The chromosomal DNA of B. parapertussis an B. bronchiseptica were igeste with the rarely cutting restriction enzyme XbaI, an the DNA fragments were resolve by PFGE. The restriction fragment patterns of these organisms were compare with each other an with the patterns of B. pertussis isolates obtaine in this an a previous stuy (6) in which an analysis of 05 isolates istinguishe 7 DNA types ientifie by PFGE on the basis of 8 to intense bans at 200 to 2. A characteristic fingerprint for B. pertussis was efine on the basis of these preominant bans. We were able to ifferentiate the three Boretella species, an each organism prouce a speciesspecific fingerprint. MATERIALS AND METHODS Isolates. A total of 25 isolates of B. pertussis (Table l), 0 isolates of B. parapertussis, an 8 isolates of B. bronchiseptica (Table 2) were examine. Isolates were obtaine from the Manchester University ollection of Bacteria (MU OB). Species were ientifie by Gram staining, cultural characteristics (B. bronchiseptica prouces visible colonies after overnight incubation on nutrient agar, B. parapertussis prouces a brown iffusible pigment after 2 ays, an B. pertussis oes not grow), an slie agglutination with rabbit polyvalent speciesspecific antiserum (, ). B. bronchiseptica was further characterize by its motility, its prouction of an obvious alkaline reaction in HughLeifson meium, an its rapi hyrolysis of urea. B. pertussis vaccine strain 700 was the control use for preparation of chromosomal DNA an molecular weight DNA markers previously size by comparison with a bacteriophage lamba DNA laer (Promega) (6). Growth of bacteria. Freezerie cultures of B. pertussis, B. parapertussis, an B. bronchiseptica were suspene in loop,l portions of sterile nutrient broth; these preparations were plate on charcoal bloo agar, nutrient agar, an bloo agar plates, respectively. The plates were incubate for 8 h 659 IP: On: Sun, 06 Jan 209 5:5:6

2 ~ 660 KHATTAK AND MATTHEWS Im. J. SYST. BATERIOL. TABLE. B. pertussis strains use in this stuy Isolate Serotype(s) Year isolate /8 2/8 / s/ D25 D252 D at 7. B. pertussis isolates were ientifie by their colonial morphology on charcoal bloo agar, Gram staining, an serotyping with monospecific agglutinating sera prepare at the Pertussis Reference Laboratory as previously escribe (). B. parapertussis isolates were ientifie on the basis of prouction of brown pigmentation on nutrient agar plates, while B. bronchiseptica prouce glistening betahemolytic colonies on bloo agar plates. A colony of each isolate was resuspene an grown for 8 h at 7 in 25 ml of moifie StainerScholte broth (5) with vigorous shaking. This synthetic meium supplemente with heptakis (2,6Oimethyl) pcycloextrin (catalog no. H679; Sigma) is a significant growth stimulant. The cells were harveste by centrifugation at,000 rpm for 0 min. The resulting pellets were resuspene in 0ml portions of 50 mm EDTA (ph 8.0), the preparations were centrifuge at,000 rpm for 0 min, an the turbiity, measure at 600 nm, was ajuste to an optical ensity of.5 in 50 mm EDTA (ph 8.0). hromosomal DNA preparations. In orer to prevent shearing of highmolecularweight DNA, ml portions of the cell suspensions escribe above were mixe with equal volumes of 2% lowtemperaturegelling molten agarose. Each cell suspensionagarose mixture was ispense into a 0plug mol (BioRa). The plugs were allowe to soliify at for 0 min. Each plug was ivie into four blocks with a clean glass coverslip. For each strain, eight blocks were incubate overnight with shaking at 7 in a sterile screwcap bottle containing 5 ml of E lysis buffer (6 mm TrisH [ph 7.5, M soium chlorie, 00 mm EDTA [ph 8.0, 0.5% [wt/vol] Brij 58 [Sigma], 0.2% [wt/vol] soium eoxycholate [BDH], 0.5% [wt/vol] Nlauryl sarcosine [Sigma], mg of lysozyme (Sigma) per ml, 20 mg of RNase [Sigma] per ml) to isrupt the cell wall an membrane. Subsequently, lysis buffer was substitute for ml of ESP solution (0.5 M EDTA [ph 9.0, % [wt/vol] Nlauryl sarcosine, mg of proteinase K per ml) for 8 h at 50 (5). The blocks were washe for at least 5 min three times in ml of 50 mm EDTA (ph 8.0) an store at until they were use. Restriction igestion of chromosomal DNA. Prior to restriction igestion of the chromosomal DNA, ialysis was necessary to eliminate the cell ebris an proteinase K activity, as well as EDTA an etergents. The presence of trace amounts of these substances interferes with restriction enonuclease igestion (6). The DNA blocks were first treate with 5 ml of TE buffer (0 mm TrisH [ph 7.5, mm EDTA [ph 8.0) containing mm phenylmethylsulfonyl fluorie for 2 h at room temperature to completely inactivate TABLE 2. B. parapertussis an B. bronchiseptica isolates use in this stuy Species Isolate Geographic location B. parapertussis MUOB 76 PPlO MIS 80 M9 MUOB 70 BR8 728DK PP5 Unite States Australia openhagen, Denmark zechoslovakia Year isolate a Serotype B. bronchiseptica MUOB 9 MUOB 2 MUOB BB MUOB openhagen, Denmark ~ ~~, unknown. IP: On: Sun, 06 Jan 209 5:5:6

3 ~ VOL., 99 GENETI RELATEDNESS OF BORDETELLA SPP. 66 DNA type TABLE. haracterization of the 2 DNA types foun in 0 isolates of B. pertusszk Sizes of DNA fragmentsa No. of isolates if if a, ban present;, ban absent;, ouble ban. Ban occurs at 25. Ban occurs at 75. Ban occurs at 297. Ban occurs at 275. f An extra ban is present at 297. * An extra ban is present at 78. Ban occurs at 2. b b g the proteinases present in the lysis solution (8). The agarose blocks were then washe three times (5 min each time) in TE buffer without phenylmethylsulfonyl fluorie. For each strain three blocks were transferre to a microcentrifuge tube an equilibrate in 00 p of XbaI reaction buffer for 20 min on ice. Subsequently, 25 U of restriction enzyme XbaI (Northumbria Biologicals) was ae, an igestion was carrie out overnight at 7. The reaction was stoppe by the aition of 0.5 ml of 50 mm EDTA (ph 8.0). PFGE. PFGE was performe with a HEFDRII system (BioRa). The gels ( by cm) were mae up of % agarose in TBE buffer (0. g of TrisH per liter, 5.5 g of boric aci per liter, 0.9 g of EDTA 0.9 per liter). The agarose blocks were loae into wells along with B. pertussis vaccine strain 700 DNA, which was use as molecular weight marker DNA. Electrophoresis was performe with a pulse time of 25 s for 0 h (6) an with a fiel strength of.5 V/cm. The gels were staine with 0.5 pg of ethiium bromie per ml in 00 ml of istille water for 20 min; this was followe by estaining for h in istille water. The gels were photographe uner a UV transilluminator. 27, 280, 5, 0, an 2. DNA type was the most common DNA type, as foun in the previous stuy (6); 7 of the 25 isolates ha this DNA type. DNA type 2 was foun in only one isolate. Four new DNA types were observe. These were esignate DNA types 8, 9, 20, an 2 (Table ). Three isolates ha DNA type 8, while two isolates ha DNA type 9. DNA types 20 an 2 were each foun in one isolate. The results for these 25 isolates are compare with the results obtaine previously (6) in Table, an the profiles of DNA types through 5 an 9 are illustrate in Fig.. Restriction igestion of the DNAs of the B. parapemssis RESULTS Macrorestriction fingerprinting by PFGE combine clear an reproucible resolution of restriction fragments with emonstration of a high egree of polymorphism. When the DNAs of 25 isolates of B. pertussis were cut withxbai, 8 to large fragments in the size range from 0 to 2 an multiple fragments smaller than 0 were generate. Fragment bans at 0 an 55 were conserve. Differentiation was base on variable bans at 200, 22, 20, 256, FIG.. B. pertussis DNA types through 5 an 9. Lanes an 7, DNA type ; lane 2, DNA type 2; lanes an 8, DNA type ; lanes an 9, DNA type ; lane 5, DNA type 5; lane 6, DNA type 9. Fragment sizes are inicate on the left. IP: On: Sun, 06 Jan 209 5:5:6

4 ~ ~~ ~ 662 KHAlTAK AND MATTHEWS INT. J. SYST. BATERIOL. FIG. 2. B. parapertussis DNA types through. Lane, molecular weight marker DNA; lanes,5, an 7 through 9, DNA type ; lanes an 6, DNA type 2; lane 2, DNA type. The sizes of fragments are inicate on the right. FIG.. B. bronchiseptica DNA types through. Lane, molecular weight marker DNA; lanes through 8, DNA type ; lane 2, DNA type 2; lane 9, DNA type. The sizes of fragments are inicate on the right. isolates withxbai generate 9 to large bans in the size range from 8 to 75. A cluster of smaller bans corresponing to sizes less than 97 long was also observe; these bans separate suboptimally uner the running conitions selecte to resolve the larger bans. Differentiation was base on variation in the four bans at 80, 200, 280, an 297. For the 0 isolates examine, three ifferent DNA types were efine on the basis of these variable bans (Fig. 2). DNA type preominate; this DNA type was foun in 7 of the 0 isolates. There were two DNA type 2 isolates an one DNA type isolate (Table ). There was no apparent correlation between phenotypic traits an genetic type. Bans at 8, 55, 66, 25, 26, 268, an 75 were conserve. When the genomes of B. brunchiseptica isolates were igeste with XbaI uner ientical conitions, up to 6 large bans in the 8 to 9 range were prouce along with multiple small bans corresponing to sizes less than 97 long. The DNA typing system was base on variation in the large bans at 8,66,200,20,20,250,256,26,268,297, an 0 (Fig. ). Five of the eight isolates of B. brunchiseptica examine ha DNA type, two isolates ha DNA type 2, an DNA type was foun in only one isolate (Table 5). There was no correlation between phenotypic an genotypic ifferences. Bans at 0,, 80, 25, an 9 were conserve. All of the isolates were typeable by this technique. The reproucibility of the technique was establishe by repeat testing, which prouce ientical fingerprint patterns uner similar running conitions. Five isolates of each species were teste at least three times, an the repeat fingerprint patterns were unchange for each of these isolates. DISUSSION The heterogeneity of the three species of the genus Buretella ientifie by phenotypic characteristics was emonstrate by the generation of speciesspecific macrorestriction profiles with the rarely cutting enzyme XbaI. The 25 isolates of B. pertussis examine in this stuy prouce six ifferent DNA profiles. These isolates were collecte between 90 an 980 (Table ). DNA type preominate in this stuy, accounting for 7 isolates. Among the seven isolates collecte before the introuction of pertussis vaccines (Table, up to 955), five were DNA type isolates. Among the remaining 8 isolates, which were collecte after the introuction of mass vaccination, 2 were DNA type isolates. Only one isolate was a DNA type 2 isolate. Four aitional DNA types were observe which were ifferent from the 7 DNA types escribe previously on the basis of an analysis of 05 isolates of B. pertussis (6). These new types were esignate DNA types 8, 9, 20, an 2 (Table ). The profiles of DNA types to 5 an 9 are shown in Fig.. No correlation was foun between the serotypes etermine by agglutination an the DNA fingerprints etermine by PFGE. All 0 isolates of B. parapertussis prouce similar DNA fingerprint patterns; 9 to large bans were prouce in the size range from 8 to 75, an 7 of these bans (8,55, 66, 25, 26, 268, an 75 ) were conserve. Differentiation within the species was base on four variable bans. The restriction igestion patterns of 7 of the 0 isolates were inistinguishable from each other an were esignate B. paraperhusis DNA type (Fig. 2 an Table ). Three of these isolates were collecte in 98, 955, an 967 from. The other four DNA type TABLE. Details of B. parapertussis DNA types recognize Sizes of DNA fragments" DNA No. of type isolates 7 2 2, ban present;, ban absent;, ouble ban. IP: On: Sun, 06 Jan 209 5:5:6

5 VOL., 99 GENETI RELATEDNESS OF BORDETELLA SPP. 66 TABLE 5. Details of B. bronchiseptica DNA types recognize Sizes of DNA fragmentsn DNA No. of type isolates 5 2 2, ban present;, ban absent;, ouble ban. isolates were from ifferent continents (Table 2). Similarly, the eight isolates of B. bronchiseptica prouce similar restriction fragment patterns (Fig. an Table 5). Unlike B. parapertussis, this species ha five conserve bans (at 0,, 80, 25 an 9 ), an ifferentiation was base on variable bans (8,66, 200,20, 20,250,256,26,268, 297, an 0 ). The restriction fragment patterns of these two species were ifferent from each other even though there was sometimes consierable variation within a species. For example, the B. bronchiseptica DNA type 2 pattern (Fig., lane 2) share just 9 of 6 bans with the B. bronchiseptica DNA type or pattern but ha only four bans in common with the B. parapertussis pattern. Bans at 9, 0, 256, 250, 20,, an 0 were foun in B. bronchiseptica DNA type 2 isolates but not in isolates of B. parapertussis. The restriction fragment patterns of these two species were compare with the restriction fragment patterns of 0 isolates of B. pertussis obtaine in this stuy an a previous stuy (6). B. pertussis ha two conserve bans (at 0 an 55 ), an ifferentiation was base on variation in 8 to intense bans between 200 an 2. There was striking interspecies iversity; none of the 0 isolates of B. pertussis showe similarity to either of the other two Boretella species (Fig. through ). In contrast, PFGE reveale consierable genetic relateness among strains of the same species. Multilocus enzyme electrophoresis reveale only a limite egree of genetic iversity among 60 strains of the three Boretella species. It was conclue that these organisms were relate too closely to warrant recognition as three separate species (0). However, multilocus enzyme electrophoresis is a measure of phenotypic expression an measures ifferences in only a few enzymes (). Arbeit et al. () emonstrate that PFGE coul ifferentiate epiemiologically inepenent but evolutionarily relate isolates of Escherichia coli that were inistinguishable by multilocus enzyme electrophoresis. Our ata confirm that macrorestriction fingerprinting by PFGE of three Boretella species can ientify consierable polymorphism among the species which was misse by multilocus enzyme electrophoresis. Similarly, the results of DNADNA hybriization stuies suggeste that these three species are closely relate (7). However, DNA samples inistinguishable on the basis of homology ata are not necessarily ientical, an DNA samples iffering by less than 5% in homology values cannot be accurately resolve (9). In conclusion, in this stuy we foun that B. pertussis, B. parapertussis, an B. bronchiseptica ha speciesspecific macrorestriction fingerprints when the organisms were examine by PFGE uner the same conitions. Isolates of the same species exhibite various egrees of similarity to each other but were easily istinguishable from the other two species. This heterogeneity in the restriction profiles of the three species, together with the phenotypic ifferences of the species, argues against the proposal of Kloos et al. (7) an Musser et al. (0) that these organisms shoul be regare as members of the same species. Macrorestriction genomic fingerprinting by PFGE reflects the overall organization of the bacterial genome an may for some species be a better inicator of clonal origin an genetic relateness than ata from other types of nucleic aci homology stuies (). It remains to be etermine whether further macrorestriction fingerprints generate by other restriction enzymes will generate some fingerprints common to two or more species of the genus Boretella. Ultimately, genomic maps, such as the one prouce for B. pertussis by Stibitz an Garletts (7), will help establish the genetic relationships of these three species. REFERENES. Arbeit, R D., M. Arthur, R. Dunn,. Kim, R. K. Selaner, an R. Golstein Resolution of recent evolutionary ivergence among Escherichiu coli from relate lineages: the application of pulsefiel electrophoresis to molecular epiemiology. J. Infect. Dis. 6:2& heesbrough, M. 98. Genus BoreteZZu, p In M. heesbrough (e.), Meical laboratory manual for tropical countries98. ambrige University Press, ambrige.. Finger, H BoreteZZu, p In S. Baron (e.), Meical microbiology986. AisonWesley Publishing o., Reaing, Mass.. Grothues, D., an EL Tummler. 99. New approaches in genome analysis by pulsefiel gel electrophoresis: application to the analysis of Pseuomonus species. Mol. Microbiol Imiazumi, A., Y. Suzuki, S. Ono, H. Sato, an Y. Sato. 98. Heptakis (2,6Oimethyl) pcycloextrin: a novel growth stimulant for BoretelZu pertussis phase I. J. lin. Microbiol. 7: Khattak, M. N., R.. Matthews, an J. P. Burnie Is BoreteZZu pertussis clonal? Br. Me. J. 0k Kloos, W. E., N. Mohapatra, W. J. Dobrogosz, J. W. Ezzell, an. R. Manclark. 98. Deoxyribonucleotie sequence relationships among Boretella species. Int. J. Syst. Bacteriol. : Mclellan, M., M. R. Jones, Y. Patel, an N. Nelson Restriction enonuclease for pulse fiel mapping of bacterial genomes. Nucleic Acis Res. 5: McFaen, J. J., an A. Knight. 99. DNA probes for etection an ientification of bacteria, p. 99. In J. M. Grange, A. Fox, an N. L. Morgan (e.), Genetic manipulation. Blackwell Scientific Publications, Oxfor. 0. Musser, J. M., E. L. Helwett, M. S. Peppler, an R. K. Selaner Genetic iversity an relationships in populations of BoreteZZu spp. J. Bacteriol. 66:2&27.. Preston, N. W Technical problems in the laboratory iagnosis an prevention of whooping cough. Lab. Pract. 9: Preston, N. W Pertussis toay, p. 8. In A.. Warlaw an R. Parton (e.), Pathogenesis an immunity in pertussis. IP: On: Sun, 06 Jan 209 5:5:6

6 66 KHATTAK AND MATTHEWS INT. J. SYST. BATERIOL. John Wiley an Sons, Lonon.. Preston, N. W Boretella: whooping cough, p In D. Greenwoo, R.. B. Slack, an J. F. Peutherer (e.), Meical microbiology. A guie to microbial infections: pathogenesis, immunity, laboratory iagnosis an control. hurchill Livingstone, Einburgh.. Sauners, N. A. 99. Analysis of restriction fragment length polymorphisms in the stuy of bacteria, p In J. M. Grange, A. Fox, an N. L. Morgan (e.), Genetic manipulation. Blackwell Scientific Publications, Oxfor. 5. Smith,. L., an. R antor Purification, specific fragmentation, an separation of large DNA molecules. Methos Enzymol. 55: Smith,. L., S. R. Kalco, an. R. antor Pulsefiel gel electrophoresis an the technology of large DNA molecules, p. 77. In R. E. Davis (e.), Genome analysis: a practical approach, st e. IRL Press, Oxfor. 7. Stibitz, S., an T. L. Garletts Derivation of a physical map of the chromosome of Boretella pertussis Tohama I. J. Bacteriol. 7: IP: On: Sun, 06 Jan 209 5:5:6