- Email: [email protected]
RAPD-PCR analysis of Staphylococcus aureus strains isolated from bovine and human hosts
ARTICLE IN PRESS Microbiological Research 159 (2004) 245–255
www.elsevier.de/micres
RAPD-PCR analysis of Staphylococcus aureus strains isolated from bovine and human hosts Elina Reinosoa, Susana Betteraa, Cecilia Frigerioa, Miguel DiRenzob, Aldo Calzolaria,c, Cristina Bognia,* Departamento de Microbiolog!ıa e Inmunolog!ıa, Facultad de Ciencias Exactas, F!ısico-Qu!ımicas y Naturales, ´rdoba, Argentina Universidad Nacional de R!ıo Cuarto, C.P. 5800 Ruta 36 km 601. R!ıo Cuarto, Co b Departamento de Mejoramiento Gene´tico, Facultad de Agronom!ıa y Veterinaria, Universidad Nacional de R!ıo ´rdoba, Argentina Cuarto, (5800) R!ıo Cuarto, Co c ´sicas, Universidad Nacional de Villa Mar!ıa, (5900) Villa Mar!ıa, Co ´rdoba, Argentina Instituto de Ciencias Ba a
Accepted 21 April 2004
KEYWORDS Staphylococcus aureus; Biotype; RAPD-PCR; Human and bovine hosts
Abstract Staphylococcus aureus is one of the most important pathogens in humans and animals. In this study eighty strains were analyzed by RAPD-PCR to assess the genetic relationship between S. aureus isolates from bovine and human hosts. Results were compared with those obtained by biotyping. Fifty-two percent of the S. aureus isolates belonged to a host specific biotype (human, bovine and poultry). Bovine and human ecovars were the most prevalent. Dendrogram obtained by RAPD results showed that all the isolates clustered into eleven groups (A–K) at a relative genetic similarity of less than 30% when analyzed with the three primers. Group A clustered 95% of the human host isolates and the remaining groups (B–K) clustered the bovine host isolates. Principal coordinate analysis also showed that the isolates could be arbitrarily divided into two groups, bovine and human, by the second coordinate. Only 9 isolates (11%) were not clustered into these groups. The genetic diversity among the S. aureus isolates from bovine hosts is relatively low compared to that of isolates from human hosts. There were no statistically significant differences among isolated from bovine and human hosts. This study shows that RAPD-PCR assayed with three primers can be successfully applied to assess the genetic relationship of S. aureus isolates from different hosts. & 2004 Elsevier GmbH. All rights reserved.
*Corresponding author. Tel./fax: þ(54)-358-4676238. E-mail address: [email protected] (C. Bogni). 0944-5013/$ - see front matter & 2004 Elsevier GmbH. All rights reserved. doi:10.1016/j.micres.2004.04.002
ARTICLE IN PRESS 246
Introduction Staphylococcus aureus is involved in a variety of invasive diseases in humans and animals. In humans it is one of the most significant pathogens causing nosocomial infections. In many countries, such as Argentina, it is the most costly disease in dairy milk production (Bogni et al., 1998; Rampone et al., 1993; Watts, 1988). On the other hand, humans act as a reservoir of S. aureus, since this bacteria may be carried in nares and hands and does not always cause disease (Williams et al., 1963). Questions have been raised to determine the relatedness of staphylococcal strain isolates from different hosts. Several studies suggest that there are differences between strains isolated from bovine and human hosts (Devriese, 1984; Ha! jek and Mars$ a! lek, 1971; Kapur et al., 1995; Musser and Kapur, 1992). Different epidemiological studies of S. aureus isolated from human and animals have been based on phenotypic techniques, such as biotyping, bacteriophage typing and antibiotic sensitivity testing (Tenover et al., 1994). Biotyping has been used as a simplified method to differentiate S. aureus isolates from humans and animals into host-specific (human, bovine, poultry and ovine) or non-host specific biotypes (Devriese, 1984). Recently, molecular DNA typing methods have been developed on the basis of genotype characterization (Tenover et al., 1994; van Belkum et al., 1995). Random amplification of polymorphic DNA (RAPD-PCR) has been applied extensively to distinguish different isolates of S. aureus (Tenover et al., 1994; Saulnier et al., 1993; van Belkum et al., 1995). However, there is little information of the RAPD method to type S. aureus strains isolated from different hosts. The aim of this work was to assess the genetic relationship of S. aureus isolates from bovine and human hosts by RAPD-PCR. Results were compared with those obtained by biotyping.
Materials and methods Bacterial isolates A total of eighty S. aureus isolates were used. Forty isolates were collected from individual quarter milk ! samples from cows of a dairy herd in Cordoba, Argentina and forty S. aureus isolates were collected from clinical episodes of disease in humans. The isolates were collected during 1999 and stored frozen at 201C with 0.8% glycerol until their use. They were cultured on brain heart
E. Reinoso et al.
infusion agar (DIFCO) and biochemically characterized as S. aureus by the standard biochemical methods described by Bair-Parker (1974).
Biotyping Biotyping was performed following the method described by Devriese (1984) based on the production of fibrinolysin, b-haemolysin, coagulation of bovine plasma and violet crystal reaction. Biotyping differentiates S. aureus isolates from human and animals into host-specific and non-host specific biotypes. The term ecovar is used by Devriese to designate biotypes which are clearly associated with a single animal host. Control strains, kindly provided by Dr. Devriese, were used as positive controls.
DNA preparation Genomic DNA was isolated as described by Janzon and Arvidson (1990). Genomic DNA was purified by ethanol precipitation and dissolved in a buffer containing 10 mM TrisClH (pH 7.6) and 0.1 mM EDTA. DNA concentration was estimated spectrophotometrically at 260 nm.
Polymerase chain reaction RAPD-PCR assay was carried out as described by Williams et al. (1990). Approximately 25 ng of chromosomal DNA was used per reaction. Amplifications were performed in 25 ml of buffer solution containing 3 mM of oligonucleotides, 200 mM of each deoxynucleoside triphosphates (Promega, Madison, USA), 3.5 mM MgCl2 and 2.5 U of DNA Taq polymerase (Promega). The oligonucleotides OLP6 (50 GAGGGAAGAG-30 ), OLP11 (50 -ACGATGAGCC-30 ) and OLP13 (50 -ACCGCCTGCT-30 ), with 60–70% of G-C content and containing non-palindromic sequences were used for DNA amplification. Mixtures were overlaid with mineral oil and amplification was performed in a PCR Express thermal cycler (Hybaid, United Kindom). The amplification consisted of a cycle of predenaturation at 941C for 5 min, followed by 40 cycles of 1 min at 931C, 1.30 min at 371C and 1 min at 721C. A final extension step of 721C for 8 min was included in all amplifications. A negative control of the same reaction mixture with water instead of chromosomal DNA was included in each run. In addition a positive control, containing the same reactive with chromosomal DNA as a well characterized reference strain (S. aureus strain ISP479) was also included. Each isolate was tested under the same conditions at least twice with the
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
selected oligonucleotides. Amplified products were separated by electrophoresis in a 1.5% agarose gel (Promega) in 0.5 X TBE buffer at a constant voltage of 4 V/cm stained with ethidium bromide (0.5 mg/ ml) and photographed with Polaroid 667 film. A 100 bp DNA marker (Promega) was used as a DNA molecular size standard.
247
software package (version 2.01e, Applied Biostatistics, Inc., New York, USA) was used. Genetic diversity was estimated by the ShannonWeaver index (Chakraborty and Rao, 1991) modified for RAPD analysis, which is defined as ! n X Ho ¼ pi log2 pi =n: i1
Pattern analysis DNA fingerprints of the isolates were compared for similarity by visual inspection of the band profiles. Subsequently, gel images were digitalized and stored as Tagget Image Files Format (TIFF). These files were converted, normalized and analyzed with GelWorks 1D software (version 3.00, Ultra Violet products, England). DNA fingerprints detected by computer were carefully verified by visual examination to correct unsatisfactory detections. Genetic relationships were established by scoring the presence or absence of each RAPD polymorphic band. The percentage of similarity between two strains was estimated by using the coefficient of Dice (1945). Cluster analysis of similarity matrices was performed by unweigthed pair group method with arithmetic averages (UPGMA) (Sneath and Sokal, 1973). When calculations were completed, a dendrogram was obtained by using the results of the RAPD assay performed with each oligonucleotide. Finally, a dendrogram using the combined results obtained with the three oligonucleotides was constructed. This combination was obtained by simply summing the number of bands (total and common) obtained in the RAPD assays with each oligonucleotide. The similarity matrices were also used as input into principal coordinate analysis (Gower, 1966). For these analyses, the NTSYS-pc
For each oligonucleotide used, Ho denotes the diversity associated with a given RAPD band; pi denotes the frequencies of ith band and n is the number of RAPD bands generated by the analyzed oligonucleotides. The Ho index varies between 0 (monomorphism) and 1 (highest polymorphism). For each host specific (bovine or human) Ho was averaged over all the oligonucleotides analyzed to determine the within host specific diversity of RAPD band (Ho). Statistical significance between origins was performed by Student’s t test (Hutcheson, 1970).
Results Biotyping of strains The results of biotyping for bovine and human isolates are shown in Table 1. All isolates were typeable by this method. Eight different biotypes (three host specific: human, bovine and poultry and five non-host specific 1–5) were identified among the bovine and human isolates. Forty-two (52.5%) of the S. aureus isolates belonged to a host specific biotype. Bovine and human ecovar were the most prevalent, representing 30% and 27.5%, respectively of all bovine isolates. The remaining bovine isolates (42.5%) were found to belong to non-host
Table 1. Classification of S. aureus into host-specific and non host-specific biotypes according to Devriese typing Biotype
Isolates from Human
Human ecovar Bovine ecovar Poultry ecovar NHS1a biotype NHS2 biotype NHS3 biotype NHS4 biotype NHS5 biotype Total a
Bovine
Total
No. of isolates
%
No. of isolates
%
No. of total isolates
%
17 0 2 3 2 2 13 1 40
42.5 0 5 7.5 5 5 32.5 2.5
11 12 0 6 7 1 1 2 40
27.5 30 0 15 17.5 2.5 2.5 5
28 12 2 9 9 3 14 3 80
35 15 2.5 11.25 11.25 3.75 17.5 3.75
NHS: Non-host-specific biotype.
ARTICLE IN PRESS 248
E. Reinoso et al.
Figure 1. RAPD-PCR profiles among S. aureus isolates of human and bovine hosts generated with each oligonucleotide. (a) OLP6, (b) OLP11 and (c) OLP13. M1: 100 bp DNA ladder. Lanes 1–5: isolates from human hosts; lanes 6-10: isolates from bovine hosts.
specific biotypes. Seventeen human isolates (42.5%) belonged to human ecovar. The remaining isolates from human hosts belonged to poultry ecovar (5%) and non-host specific biotypes (52.5%). No human isolates were associated to bovine ecovar. Non-host specific biotype 4 was the most frequent, with a prevalence of 32.5%.
RAPD-PCR profile In this study, RAPD-PCR was used as a tool to assess the genetic relationship of eighty S. aureus strains isolated from bovine and human hosts. The three oligonucleotides generated a high number of bands and a limited number of low intensity bands. A set of reproducible bands produced for a particular oligonucleotide was defined as ‘‘profile’’. RAPD profile resulted in DNA amplification fragments ranged in size from 300 to 1900 bp. Fig. 1 shows the results obtained with each oligonucleotide. The number of amplified fragments ranged from 1 to 14 and high molecular weight bands (among 1500 and 2900 bp) were shared by the human strains. RAPD profiles obtained with oligonucleotides OLP6, OLP11 and OLP13 were 47, 67 and 68, respectively. Dendrograms constructed with each oligonucleotide showed that several isolates could not be distinguished from each other (Fig. 2a–c). In addition, the strains that clustered together sometimes differed, depending on which oligonucleotide was used. When calculations were made by using the summed results for the three oligonucleotides,
the number of different profiles increased to 79 (Fig. 3). Only strains 618 and 619, isolated from the bovine hosts, clustered together. The eighty S. aureus isolated were divided into eleven major groups (A–K), separated at a relative genetic similarity of less than 30%. Group A clustered 95% of human host isolates, only two strains (603 and 616) isolated from the bovine host remained in this group. Bovine isolates clustered into 10 groups (B– K) and strains 816 and 854 isolated from human hosts remained in group B and C, respectively. Principal coordinate analysis showed that the isolates were divided into two groups, bovine and human, by the second coordinate (Fig. 4). Only nine S. aureus isolates (11%) did not clustered into these groups. Genetic diversity within bovine and human S. aureus isolates calculated by Shannon-Weaver index, with the three oligonucleotides showed that the average diversity (Ho) for human and bovine origin was 0.56 and 0.36, respectively. When groups were compared by the t test, significant differences were not found (P40:05).
Discussion The occurrence of host specificity among bacterial isolates has been described for a variety of pathogenic bacteria (Selander and Musser, 1990; Jers$ ek et al., 1999; Mart!ınez et al., 2000). The concept that S. aureus strains isolated from different hosts differ from each other was first
Figure 2. Genetic relationship between 80 S. aureus isolates, from human and bovine hosts, as estimated by clustering analysis of RAPD profiles obtained with each oligonucleotide. The dendrograms were generated by the unweighted pair group method with arithmetic averages (UPGMA). (a) Oligonucleotide OLP6, (b) Oligonucleotide OLP11, and (c) Oligonucleotide OLP13.
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
249
ARTICLE IN PRESS 250
E. Reinoso et al.
Figure 2. (continued).
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
Figure 2. (continued).
251
ARTICLE IN PRESS 252
E. Reinoso et al.
Figure 3. Genetic relationship between 80 S. aureus isolates, from human and bovine hosts, as estimated by clustering analysis of RAPD profiles obtained with the three oligonucleotides. The dendrogram was generated by the unweighted pair group method with arithmetic averages (UPGMA).
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
253
Figure 4. Ordination of S. aureus isolates based on the principal component analysis obtained with the combination of the three primers. H ¼ S. aureus isolated from human hosts, B ¼ S. aureus isolated from bovine hosts.
appreciated by Madison (1935) many years ago. Several reports demonstrated that the origin of S. aureus isolates can be traced with the help of a few tests (Ha! jek and Marsa! lek, 1971; Devriese, 1984). Formal propositions for subdivision of S. aureus into
biotypes, varieties or ecovars, according to the natural host origin, were made by different researchers (Devriese, 1984; Ha! jek and Marsa! lek, 1971). Devriese biotyping has been successfully used in ecological and epidemiological studies of
ARTICLE IN PRESS 254
S. aureus for many years and provides a method to differentiate isolates from human and animals into host and non-host-specific biotypes. Modern typing methods based on genotype characterization also attempt to associate bacteria with host-specificity (Selander and Musser, 1990; Kapur et al., 1995; Zadoks et al., 2000). In this study, only 42 of 80 S. aureus isolates (52.5%) assayed by biotyping could be associated to a host specific biotype. The other isolates (47.5%) belonged to five non-specific biotypes. Bovine ecovar was not found among human isolates, and S. aureus isolates belonged to poultry ecovar were collected from human hosts, as was observed by Devriese (1984). When RAPD-PCR profiles were analyzed, 90% were preferentially associated with a single host. A dendrogram constructed on the basis of RAPD-PCR profile with the three oligonucleotides, showed the presence of 79 profiles. The eighty isolates were divided into eleven groups (A– K), most of the human isolates belonged to group A and bovine isolates were distributed over the other ten groups. This level of clustering was based on the origin of S. aureus strains (bovine or human). Principal coordinate analysis showed that the isolates were divided into two groups, bovine and human, by the second coordinate. Taken together, these results are consistent with the concept of host specialization as was reported by Kreiswirth et al. (1989), who observed that TSST-1 producing S. aureus isolates recovered from bovine hosts are genotypically distinct from the major human TSST-1 isolates. The genetic diversity of 0.36 among the S. aureus isolates from bovine hosts is relatively low compared with those isolates from human hosts. These results are in agreement with Kapur et al. (1995), who reported that transfer of S. aureus between bovine and human hosts is limited when the isolates were analyzed by multilocus enzyme electrophoresis. However, several studies suggest that transfer of bacteria between humans and animals is possible and have implicated milker’s hands in the spread of S. aureus strains associated with bovine intramammary infections (Fox et al., 1991; Aarestrup et al., 1995). This study shows that RAPD-PCR can be successfully applied to assess the genetic relationship of S. aureus isolates from different hosts.
Acknowledgements This work was supported by grants from Consejo de ! Investigaciones Cient!ıficas y Tecnologicas de la
E. Reinoso et al. Provincia de Co! rdoba and Secretar!ıa de Ciencia y Te ´cnica de la Universidad Nacional de R!ıo Cuarto.
References Aarestrup, F., Wegener, H., Rosdahl, V., 1995. Evaluation of phenotypic and genotypic methods for epidemiological typing of Staphylococcus aureus isolates from bovine mastitis in Denmark. Vet. Microbiol. 45, 139–150. Bair-Parker, A.C., 1974. Micrococcaceae. In: Buchanan, R.E., Gibbons, N.E. (Eds.), Bergey’s manual of determinative bacteriology. 8th ed., Baltimore, USA, pp. 478–490. Bogni, C., Calzolari, A., Nagel, R., Segura, M., Giraudo, A., Giraudo, J., 1998. Avirulent and inmunogenecity in mice of a bovine mastitis S. aureus. Can. J. Vet. Res. 62, 235–239. Chakraborty, R., Rao, C.R., 1991. Measurements of genetic variation for evolutionary studies. In: Rao, C.R., Chakraborty, R. (Eds.), Handbook of statistics, Vol. 8. Elsevier, Amsterdam, Holand, pp. 271–316. Devriese, L.A., 1984. A simplified system for biotyping Staphylococcus aureus strains isolated from different animal species. J. Appl. Bacteriol. 56, 215–220. Dice, L.R., 1945. Measurements of the amount of ecologic association between species. Ecology 26, 297–302. Fox, L.K., Gershman, M., Hanock, D.D., Hutton, C.T., 1991. Fomites and reservoirs of Staphylococcus aureus causing intrammary infections as determined by phage typing: the effect of milking time hygiene practices. Cornell Vet. 81, 183–193. Gower, J.C., 1966. Some distance properties of latent root and vector methods using multivariate analysis. Biometrika 53, 325–338. Ha! jek, V., Marsa! lek, E., 1971. The differentiation of pathogenic staphylococci and a suggestion for their taxonomic classification. Zentralbl. Bakteriol. Abt. I. Orig. A 217, 176–182. Hutcheson, K., 1970. A test for comparing diversities based on the Shannon Formula. J. Theor. Biol. 29, 151–154. Janzon, L., Arvidson, S., 1990. The role of the delta-lysin gene (hld) in the regulation of virulence genes by the accessory gene regulator (agr) in Staphylococcus aureus. EMBO J. 9, 1391–1399. Jers$ ek, B., Gilot, P., Gubina, M., Klun, N., Mehle, J., Tcherneva, E., Rijpens, N., Herman, L., 1999. Typing of Listeria monocytogenes strains by repetitive element sequence-based PCR. J. Clin. Microbiol. 37, 103–109. Kapur, V., Sischo, W., Greer, R., Whittamh, T., Musser, J., 1995. Molecular population genetic analysis of the S. aureus recovered from cows. J. Clin. Microbiol. 33, 376–380. Kreiswirth, B.N., Projan, S.J., Schlievert, P.M., Novick, R.P., 1989. Toxic shock syndrome toxin 1 is encoded by a variable genetic element. Rev Infect Dis. 1, 83–88.
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
Madison, R.R., 1935. Fibrinolytic staphylococci. Proc. Soc. Biol. Med. 33, 210–211. Mart!ınez, G., Harel, J., Higgins, R., Lacoutore, S., Daignault, D., Gottschalk, M., 2000. Characterization of Streptococcus agalactiae isolates from bovine and human origin by ramdomly amplified polymorphic DNA analysis (RAPD). J. Clin. Microbiol. 38, 71–78. Musser, J., Kapur, V., 1992. Clonal analysis of methicillin resistant Staphylococcus aureus strains from intercontinental sources: association of the mec gene with divergent phylogenetic lineages implies dissemination by horizontal transfer and recombination. J. Clin. Microbiol. 30, 2058–2063. Rampone, H., Giraudo, J., Bogni, C., Calzolari, A., 1993. Frequency of isolation of coagulase negative staphylococci in normal and mastitis milk. Zentralbl. Bakteriol. 279, 537–543. Saulnier, P., Bourneix, C., Pre ´vost, G., Andremont, A., 1993. Random amplified polymorphic DNA assay is less discriminatory than pulsed field gel electrophoresis for typing strains of methicillin resistant Staphylococcus aureus. J. Clin. Microbiol. 31, 982–985. Selander, R., Musser, J., 1990. Population genetics of bacterial pathogenesis. In: Iglewski, B.H., Clark, V. (Eds.), Molecular basis of bacterial pathogenesis. Academic Press, San Diego, USA, pp. 11–36. Sneath, P., Sokal, R., 1973. In: Freeman, W. H. (Ed.), Numerical taxonomy: the principles and practice of
255
numerical classification. W. H. Frecman and co, San Francisco, USA. Tenover, F., Arbeit, R., Archer, G., Biddle, J., Byrne, S., Goering, R., Hancock, G., He ´bert, A., Hill, B., Hollis, R., Jarvis, W., Kreiswirth, B., Eisner, W., Maslow, J., McDougal, L., Miller, M., Mulligan, M., Pfaller, M., 1994. Comparison of traditional and molecular methods of typing isolated of Staphylococcus aureus. J. Clin. Microbiol. 32, 407–415. Van Belkum, A., Kluytamans, J., Van Leeuwen, W., Bax, R., Quint, W., Peters, E., Fluit, A., VandenbrouckeGrauls, C., Van den Brule, A., Koelman, H., 1995. Multicenter evaluation of arbitrarily primed PCR for typing of Staphylococcus aureus strains. J. Clin. Microbiol. 33, 1537–1547. Watts, J., 1988. Etiological agents of bovine mastitis. Vet. Microbiol. 16, 41–46. Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A., Tingey, S.V., 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucl. Acids Res. 18, 6531–6535. Zadoks, R., van Leeuwen, W., Barkema, H., Sampimon, O., Verbrugh, H., Schukken, Y.H., van Belkum, A., 2000. Application of pulsed-field gel electrophoresis and binary typing as tools in veterinary clinical microbiology and molecular epidemiologic analysis of bovine and human Staphylococcus aureus isolates. J. Clin. Microbiol. 38, 1931–1939.
www.elsevier.de/micres
RAPD-PCR analysis of Staphylococcus aureus strains isolated from bovine and human hosts Elina Reinosoa, Susana Betteraa, Cecilia Frigerioa, Miguel DiRenzob, Aldo Calzolaria,c, Cristina Bognia,* Departamento de Microbiolog!ıa e Inmunolog!ıa, Facultad de Ciencias Exactas, F!ısico-Qu!ımicas y Naturales, ´rdoba, Argentina Universidad Nacional de R!ıo Cuarto, C.P. 5800 Ruta 36 km 601. R!ıo Cuarto, Co b Departamento de Mejoramiento Gene´tico, Facultad de Agronom!ıa y Veterinaria, Universidad Nacional de R!ıo ´rdoba, Argentina Cuarto, (5800) R!ıo Cuarto, Co c ´sicas, Universidad Nacional de Villa Mar!ıa, (5900) Villa Mar!ıa, Co ´rdoba, Argentina Instituto de Ciencias Ba a
Accepted 21 April 2004
KEYWORDS Staphylococcus aureus; Biotype; RAPD-PCR; Human and bovine hosts
Abstract Staphylococcus aureus is one of the most important pathogens in humans and animals. In this study eighty strains were analyzed by RAPD-PCR to assess the genetic relationship between S. aureus isolates from bovine and human hosts. Results were compared with those obtained by biotyping. Fifty-two percent of the S. aureus isolates belonged to a host specific biotype (human, bovine and poultry). Bovine and human ecovars were the most prevalent. Dendrogram obtained by RAPD results showed that all the isolates clustered into eleven groups (A–K) at a relative genetic similarity of less than 30% when analyzed with the three primers. Group A clustered 95% of the human host isolates and the remaining groups (B–K) clustered the bovine host isolates. Principal coordinate analysis also showed that the isolates could be arbitrarily divided into two groups, bovine and human, by the second coordinate. Only 9 isolates (11%) were not clustered into these groups. The genetic diversity among the S. aureus isolates from bovine hosts is relatively low compared to that of isolates from human hosts. There were no statistically significant differences among isolated from bovine and human hosts. This study shows that RAPD-PCR assayed with three primers can be successfully applied to assess the genetic relationship of S. aureus isolates from different hosts. & 2004 Elsevier GmbH. All rights reserved.
*Corresponding author. Tel./fax: þ(54)-358-4676238. E-mail address: [email protected] (C. Bogni). 0944-5013/$ - see front matter & 2004 Elsevier GmbH. All rights reserved. doi:10.1016/j.micres.2004.04.002
ARTICLE IN PRESS 246
Introduction Staphylococcus aureus is involved in a variety of invasive diseases in humans and animals. In humans it is one of the most significant pathogens causing nosocomial infections. In many countries, such as Argentina, it is the most costly disease in dairy milk production (Bogni et al., 1998; Rampone et al., 1993; Watts, 1988). On the other hand, humans act as a reservoir of S. aureus, since this bacteria may be carried in nares and hands and does not always cause disease (Williams et al., 1963). Questions have been raised to determine the relatedness of staphylococcal strain isolates from different hosts. Several studies suggest that there are differences between strains isolated from bovine and human hosts (Devriese, 1984; Ha! jek and Mars$ a! lek, 1971; Kapur et al., 1995; Musser and Kapur, 1992). Different epidemiological studies of S. aureus isolated from human and animals have been based on phenotypic techniques, such as biotyping, bacteriophage typing and antibiotic sensitivity testing (Tenover et al., 1994). Biotyping has been used as a simplified method to differentiate S. aureus isolates from humans and animals into host-specific (human, bovine, poultry and ovine) or non-host specific biotypes (Devriese, 1984). Recently, molecular DNA typing methods have been developed on the basis of genotype characterization (Tenover et al., 1994; van Belkum et al., 1995). Random amplification of polymorphic DNA (RAPD-PCR) has been applied extensively to distinguish different isolates of S. aureus (Tenover et al., 1994; Saulnier et al., 1993; van Belkum et al., 1995). However, there is little information of the RAPD method to type S. aureus strains isolated from different hosts. The aim of this work was to assess the genetic relationship of S. aureus isolates from bovine and human hosts by RAPD-PCR. Results were compared with those obtained by biotyping.
Materials and methods Bacterial isolates A total of eighty S. aureus isolates were used. Forty isolates were collected from individual quarter milk ! samples from cows of a dairy herd in Cordoba, Argentina and forty S. aureus isolates were collected from clinical episodes of disease in humans. The isolates were collected during 1999 and stored frozen at 201C with 0.8% glycerol until their use. They were cultured on brain heart
E. Reinoso et al.
infusion agar (DIFCO) and biochemically characterized as S. aureus by the standard biochemical methods described by Bair-Parker (1974).
Biotyping Biotyping was performed following the method described by Devriese (1984) based on the production of fibrinolysin, b-haemolysin, coagulation of bovine plasma and violet crystal reaction. Biotyping differentiates S. aureus isolates from human and animals into host-specific and non-host specific biotypes. The term ecovar is used by Devriese to designate biotypes which are clearly associated with a single animal host. Control strains, kindly provided by Dr. Devriese, were used as positive controls.
DNA preparation Genomic DNA was isolated as described by Janzon and Arvidson (1990). Genomic DNA was purified by ethanol precipitation and dissolved in a buffer containing 10 mM TrisClH (pH 7.6) and 0.1 mM EDTA. DNA concentration was estimated spectrophotometrically at 260 nm.
Polymerase chain reaction RAPD-PCR assay was carried out as described by Williams et al. (1990). Approximately 25 ng of chromosomal DNA was used per reaction. Amplifications were performed in 25 ml of buffer solution containing 3 mM of oligonucleotides, 200 mM of each deoxynucleoside triphosphates (Promega, Madison, USA), 3.5 mM MgCl2 and 2.5 U of DNA Taq polymerase (Promega). The oligonucleotides OLP6 (50 GAGGGAAGAG-30 ), OLP11 (50 -ACGATGAGCC-30 ) and OLP13 (50 -ACCGCCTGCT-30 ), with 60–70% of G-C content and containing non-palindromic sequences were used for DNA amplification. Mixtures were overlaid with mineral oil and amplification was performed in a PCR Express thermal cycler (Hybaid, United Kindom). The amplification consisted of a cycle of predenaturation at 941C for 5 min, followed by 40 cycles of 1 min at 931C, 1.30 min at 371C and 1 min at 721C. A final extension step of 721C for 8 min was included in all amplifications. A negative control of the same reaction mixture with water instead of chromosomal DNA was included in each run. In addition a positive control, containing the same reactive with chromosomal DNA as a well characterized reference strain (S. aureus strain ISP479) was also included. Each isolate was tested under the same conditions at least twice with the
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
selected oligonucleotides. Amplified products were separated by electrophoresis in a 1.5% agarose gel (Promega) in 0.5 X TBE buffer at a constant voltage of 4 V/cm stained with ethidium bromide (0.5 mg/ ml) and photographed with Polaroid 667 film. A 100 bp DNA marker (Promega) was used as a DNA molecular size standard.
247
software package (version 2.01e, Applied Biostatistics, Inc., New York, USA) was used. Genetic diversity was estimated by the ShannonWeaver index (Chakraborty and Rao, 1991) modified for RAPD analysis, which is defined as ! n X Ho ¼ pi log2 pi =n: i1
Pattern analysis DNA fingerprints of the isolates were compared for similarity by visual inspection of the band profiles. Subsequently, gel images were digitalized and stored as Tagget Image Files Format (TIFF). These files were converted, normalized and analyzed with GelWorks 1D software (version 3.00, Ultra Violet products, England). DNA fingerprints detected by computer were carefully verified by visual examination to correct unsatisfactory detections. Genetic relationships were established by scoring the presence or absence of each RAPD polymorphic band. The percentage of similarity between two strains was estimated by using the coefficient of Dice (1945). Cluster analysis of similarity matrices was performed by unweigthed pair group method with arithmetic averages (UPGMA) (Sneath and Sokal, 1973). When calculations were completed, a dendrogram was obtained by using the results of the RAPD assay performed with each oligonucleotide. Finally, a dendrogram using the combined results obtained with the three oligonucleotides was constructed. This combination was obtained by simply summing the number of bands (total and common) obtained in the RAPD assays with each oligonucleotide. The similarity matrices were also used as input into principal coordinate analysis (Gower, 1966). For these analyses, the NTSYS-pc
For each oligonucleotide used, Ho denotes the diversity associated with a given RAPD band; pi denotes the frequencies of ith band and n is the number of RAPD bands generated by the analyzed oligonucleotides. The Ho index varies between 0 (monomorphism) and 1 (highest polymorphism). For each host specific (bovine or human) Ho was averaged over all the oligonucleotides analyzed to determine the within host specific diversity of RAPD band (Ho). Statistical significance between origins was performed by Student’s t test (Hutcheson, 1970).
Results Biotyping of strains The results of biotyping for bovine and human isolates are shown in Table 1. All isolates were typeable by this method. Eight different biotypes (three host specific: human, bovine and poultry and five non-host specific 1–5) were identified among the bovine and human isolates. Forty-two (52.5%) of the S. aureus isolates belonged to a host specific biotype. Bovine and human ecovar were the most prevalent, representing 30% and 27.5%, respectively of all bovine isolates. The remaining bovine isolates (42.5%) were found to belong to non-host
Table 1. Classification of S. aureus into host-specific and non host-specific biotypes according to Devriese typing Biotype
Isolates from Human
Human ecovar Bovine ecovar Poultry ecovar NHS1a biotype NHS2 biotype NHS3 biotype NHS4 biotype NHS5 biotype Total a
Bovine
Total
No. of isolates
%
No. of isolates
%
No. of total isolates
%
17 0 2 3 2 2 13 1 40
42.5 0 5 7.5 5 5 32.5 2.5
11 12 0 6 7 1 1 2 40
27.5 30 0 15 17.5 2.5 2.5 5
28 12 2 9 9 3 14 3 80
35 15 2.5 11.25 11.25 3.75 17.5 3.75
NHS: Non-host-specific biotype.
ARTICLE IN PRESS 248
E. Reinoso et al.
Figure 1. RAPD-PCR profiles among S. aureus isolates of human and bovine hosts generated with each oligonucleotide. (a) OLP6, (b) OLP11 and (c) OLP13. M1: 100 bp DNA ladder. Lanes 1–5: isolates from human hosts; lanes 6-10: isolates from bovine hosts.
specific biotypes. Seventeen human isolates (42.5%) belonged to human ecovar. The remaining isolates from human hosts belonged to poultry ecovar (5%) and non-host specific biotypes (52.5%). No human isolates were associated to bovine ecovar. Non-host specific biotype 4 was the most frequent, with a prevalence of 32.5%.
RAPD-PCR profile In this study, RAPD-PCR was used as a tool to assess the genetic relationship of eighty S. aureus strains isolated from bovine and human hosts. The three oligonucleotides generated a high number of bands and a limited number of low intensity bands. A set of reproducible bands produced for a particular oligonucleotide was defined as ‘‘profile’’. RAPD profile resulted in DNA amplification fragments ranged in size from 300 to 1900 bp. Fig. 1 shows the results obtained with each oligonucleotide. The number of amplified fragments ranged from 1 to 14 and high molecular weight bands (among 1500 and 2900 bp) were shared by the human strains. RAPD profiles obtained with oligonucleotides OLP6, OLP11 and OLP13 were 47, 67 and 68, respectively. Dendrograms constructed with each oligonucleotide showed that several isolates could not be distinguished from each other (Fig. 2a–c). In addition, the strains that clustered together sometimes differed, depending on which oligonucleotide was used. When calculations were made by using the summed results for the three oligonucleotides,
the number of different profiles increased to 79 (Fig. 3). Only strains 618 and 619, isolated from the bovine hosts, clustered together. The eighty S. aureus isolated were divided into eleven major groups (A–K), separated at a relative genetic similarity of less than 30%. Group A clustered 95% of human host isolates, only two strains (603 and 616) isolated from the bovine host remained in this group. Bovine isolates clustered into 10 groups (B– K) and strains 816 and 854 isolated from human hosts remained in group B and C, respectively. Principal coordinate analysis showed that the isolates were divided into two groups, bovine and human, by the second coordinate (Fig. 4). Only nine S. aureus isolates (11%) did not clustered into these groups. Genetic diversity within bovine and human S. aureus isolates calculated by Shannon-Weaver index, with the three oligonucleotides showed that the average diversity (Ho) for human and bovine origin was 0.56 and 0.36, respectively. When groups were compared by the t test, significant differences were not found (P40:05).
Discussion The occurrence of host specificity among bacterial isolates has been described for a variety of pathogenic bacteria (Selander and Musser, 1990; Jers$ ek et al., 1999; Mart!ınez et al., 2000). The concept that S. aureus strains isolated from different hosts differ from each other was first
Figure 2. Genetic relationship between 80 S. aureus isolates, from human and bovine hosts, as estimated by clustering analysis of RAPD profiles obtained with each oligonucleotide. The dendrograms were generated by the unweighted pair group method with arithmetic averages (UPGMA). (a) Oligonucleotide OLP6, (b) Oligonucleotide OLP11, and (c) Oligonucleotide OLP13.
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
249
ARTICLE IN PRESS 250
E. Reinoso et al.
Figure 2. (continued).
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
Figure 2. (continued).
251
ARTICLE IN PRESS 252
E. Reinoso et al.
Figure 3. Genetic relationship between 80 S. aureus isolates, from human and bovine hosts, as estimated by clustering analysis of RAPD profiles obtained with the three oligonucleotides. The dendrogram was generated by the unweighted pair group method with arithmetic averages (UPGMA).
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
253
Figure 4. Ordination of S. aureus isolates based on the principal component analysis obtained with the combination of the three primers. H ¼ S. aureus isolated from human hosts, B ¼ S. aureus isolated from bovine hosts.
appreciated by Madison (1935) many years ago. Several reports demonstrated that the origin of S. aureus isolates can be traced with the help of a few tests (Ha! jek and Marsa! lek, 1971; Devriese, 1984). Formal propositions for subdivision of S. aureus into
biotypes, varieties or ecovars, according to the natural host origin, were made by different researchers (Devriese, 1984; Ha! jek and Marsa! lek, 1971). Devriese biotyping has been successfully used in ecological and epidemiological studies of
ARTICLE IN PRESS 254
S. aureus for many years and provides a method to differentiate isolates from human and animals into host and non-host-specific biotypes. Modern typing methods based on genotype characterization also attempt to associate bacteria with host-specificity (Selander and Musser, 1990; Kapur et al., 1995; Zadoks et al., 2000). In this study, only 42 of 80 S. aureus isolates (52.5%) assayed by biotyping could be associated to a host specific biotype. The other isolates (47.5%) belonged to five non-specific biotypes. Bovine ecovar was not found among human isolates, and S. aureus isolates belonged to poultry ecovar were collected from human hosts, as was observed by Devriese (1984). When RAPD-PCR profiles were analyzed, 90% were preferentially associated with a single host. A dendrogram constructed on the basis of RAPD-PCR profile with the three oligonucleotides, showed the presence of 79 profiles. The eighty isolates were divided into eleven groups (A– K), most of the human isolates belonged to group A and bovine isolates were distributed over the other ten groups. This level of clustering was based on the origin of S. aureus strains (bovine or human). Principal coordinate analysis showed that the isolates were divided into two groups, bovine and human, by the second coordinate. Taken together, these results are consistent with the concept of host specialization as was reported by Kreiswirth et al. (1989), who observed that TSST-1 producing S. aureus isolates recovered from bovine hosts are genotypically distinct from the major human TSST-1 isolates. The genetic diversity of 0.36 among the S. aureus isolates from bovine hosts is relatively low compared with those isolates from human hosts. These results are in agreement with Kapur et al. (1995), who reported that transfer of S. aureus between bovine and human hosts is limited when the isolates were analyzed by multilocus enzyme electrophoresis. However, several studies suggest that transfer of bacteria between humans and animals is possible and have implicated milker’s hands in the spread of S. aureus strains associated with bovine intramammary infections (Fox et al., 1991; Aarestrup et al., 1995). This study shows that RAPD-PCR can be successfully applied to assess the genetic relationship of S. aureus isolates from different hosts.
Acknowledgements This work was supported by grants from Consejo de ! Investigaciones Cient!ıficas y Tecnologicas de la
E. Reinoso et al. Provincia de Co! rdoba and Secretar!ıa de Ciencia y Te ´cnica de la Universidad Nacional de R!ıo Cuarto.
References Aarestrup, F., Wegener, H., Rosdahl, V., 1995. Evaluation of phenotypic and genotypic methods for epidemiological typing of Staphylococcus aureus isolates from bovine mastitis in Denmark. Vet. Microbiol. 45, 139–150. Bair-Parker, A.C., 1974. Micrococcaceae. In: Buchanan, R.E., Gibbons, N.E. (Eds.), Bergey’s manual of determinative bacteriology. 8th ed., Baltimore, USA, pp. 478–490. Bogni, C., Calzolari, A., Nagel, R., Segura, M., Giraudo, A., Giraudo, J., 1998. Avirulent and inmunogenecity in mice of a bovine mastitis S. aureus. Can. J. Vet. Res. 62, 235–239. Chakraborty, R., Rao, C.R., 1991. Measurements of genetic variation for evolutionary studies. In: Rao, C.R., Chakraborty, R. (Eds.), Handbook of statistics, Vol. 8. Elsevier, Amsterdam, Holand, pp. 271–316. Devriese, L.A., 1984. A simplified system for biotyping Staphylococcus aureus strains isolated from different animal species. J. Appl. Bacteriol. 56, 215–220. Dice, L.R., 1945. Measurements of the amount of ecologic association between species. Ecology 26, 297–302. Fox, L.K., Gershman, M., Hanock, D.D., Hutton, C.T., 1991. Fomites and reservoirs of Staphylococcus aureus causing intrammary infections as determined by phage typing: the effect of milking time hygiene practices. Cornell Vet. 81, 183–193. Gower, J.C., 1966. Some distance properties of latent root and vector methods using multivariate analysis. Biometrika 53, 325–338. Ha! jek, V., Marsa! lek, E., 1971. The differentiation of pathogenic staphylococci and a suggestion for their taxonomic classification. Zentralbl. Bakteriol. Abt. I. Orig. A 217, 176–182. Hutcheson, K., 1970. A test for comparing diversities based on the Shannon Formula. J. Theor. Biol. 29, 151–154. Janzon, L., Arvidson, S., 1990. The role of the delta-lysin gene (hld) in the regulation of virulence genes by the accessory gene regulator (agr) in Staphylococcus aureus. EMBO J. 9, 1391–1399. Jers$ ek, B., Gilot, P., Gubina, M., Klun, N., Mehle, J., Tcherneva, E., Rijpens, N., Herman, L., 1999. Typing of Listeria monocytogenes strains by repetitive element sequence-based PCR. J. Clin. Microbiol. 37, 103–109. Kapur, V., Sischo, W., Greer, R., Whittamh, T., Musser, J., 1995. Molecular population genetic analysis of the S. aureus recovered from cows. J. Clin. Microbiol. 33, 376–380. Kreiswirth, B.N., Projan, S.J., Schlievert, P.M., Novick, R.P., 1989. Toxic shock syndrome toxin 1 is encoded by a variable genetic element. Rev Infect Dis. 1, 83–88.
ARTICLE IN PRESS RAPD-PCR analysis of staphylococcus aureus strains isolated from bovine and human hosts
Madison, R.R., 1935. Fibrinolytic staphylococci. Proc. Soc. Biol. Med. 33, 210–211. Mart!ınez, G., Harel, J., Higgins, R., Lacoutore, S., Daignault, D., Gottschalk, M., 2000. Characterization of Streptococcus agalactiae isolates from bovine and human origin by ramdomly amplified polymorphic DNA analysis (RAPD). J. Clin. Microbiol. 38, 71–78. Musser, J., Kapur, V., 1992. Clonal analysis of methicillin resistant Staphylococcus aureus strains from intercontinental sources: association of the mec gene with divergent phylogenetic lineages implies dissemination by horizontal transfer and recombination. J. Clin. Microbiol. 30, 2058–2063. Rampone, H., Giraudo, J., Bogni, C., Calzolari, A., 1993. Frequency of isolation of coagulase negative staphylococci in normal and mastitis milk. Zentralbl. Bakteriol. 279, 537–543. Saulnier, P., Bourneix, C., Pre ´vost, G., Andremont, A., 1993. Random amplified polymorphic DNA assay is less discriminatory than pulsed field gel electrophoresis for typing strains of methicillin resistant Staphylococcus aureus. J. Clin. Microbiol. 31, 982–985. Selander, R., Musser, J., 1990. Population genetics of bacterial pathogenesis. In: Iglewski, B.H., Clark, V. (Eds.), Molecular basis of bacterial pathogenesis. Academic Press, San Diego, USA, pp. 11–36. Sneath, P., Sokal, R., 1973. In: Freeman, W. H. (Ed.), Numerical taxonomy: the principles and practice of
255
numerical classification. W. H. Frecman and co, San Francisco, USA. Tenover, F., Arbeit, R., Archer, G., Biddle, J., Byrne, S., Goering, R., Hancock, G., He ´bert, A., Hill, B., Hollis, R., Jarvis, W., Kreiswirth, B., Eisner, W., Maslow, J., McDougal, L., Miller, M., Mulligan, M., Pfaller, M., 1994. Comparison of traditional and molecular methods of typing isolated of Staphylococcus aureus. J. Clin. Microbiol. 32, 407–415. Van Belkum, A., Kluytamans, J., Van Leeuwen, W., Bax, R., Quint, W., Peters, E., Fluit, A., VandenbrouckeGrauls, C., Van den Brule, A., Koelman, H., 1995. Multicenter evaluation of arbitrarily primed PCR for typing of Staphylococcus aureus strains. J. Clin. Microbiol. 33, 1537–1547. Watts, J., 1988. Etiological agents of bovine mastitis. Vet. Microbiol. 16, 41–46. Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A., Tingey, S.V., 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucl. Acids Res. 18, 6531–6535. Zadoks, R., van Leeuwen, W., Barkema, H., Sampimon, O., Verbrugh, H., Schukken, Y.H., van Belkum, A., 2000. Application of pulsed-field gel electrophoresis and binary typing as tools in veterinary clinical microbiology and molecular epidemiologic analysis of bovine and human Staphylococcus aureus isolates. J. Clin. Microbiol. 38, 1931–1939.