Determining the Genetic Structure of the Natural Population of Staphylococcus aureus: a Comparison of Multilocus Sequence Typing with Pulsed-Field Gel Electrophoresis, Randomly Amplified Polymorphic DNA Analysis, and Phage Typing

We used a sample of Staphylococcus aureus strains that are carriedby humans and that are representative of the natural populationof S. aureus strains in order to assess the value of multilocussequence typing (MLST), pulsed-field gel electrophoresis, randomlyamplified polymorphic DNA analysis, and phage typing as epidemiologicaltools. Only MLST was able to define clonal complexes unambiguously.All DNA-based typing approaches achieved a high degree of agreement,implying phylogenetic concordance, but predicted epidemiologicalassociations with variable accuracy.

INTRODUCTION

Top

Introduction

Staphylococcus aureus is one of the most important pathogensin clinical settings. It is also one of the leading causes ofnosocomial infections and the dissemination of multiple-drug-resistantstrains, mainly methicillin-resistant S. aureus (MRSA), andthe recent emergence of a vancomycin-resistant MRSA is of concernto hospitals worldwide. Naturally, an understanding of the dynamicsof spread and an identification of transmissions or outbreaksare of interest not only to public health epidemiologists butalso to clinical microbiologists and indeed clinicians involvedin patient management on a daily basis. Most of the currentepidemiological typing schemes compare fragment patterns generatedby restriction or amplification of chromosomal DNA or by differencesin lysis patterns based on susceptibilities to lytic bacteriophages.All of these approaches are able to distinguish between unrelatedstrains (albeit with different precision), but the probabilitywith which two indistinguishable patterns predict an epidemiologicalrelationship, i.e., transmission, cannot be determined withoutknowledge of the underlying genetic structure of the naturallyoccurring population. In order to describe the genetic structureof bacterial populations and the abundance of certain clonesin a given environment, typing tools such as multilocus sequencetyping (MLST) provide a comprehensive genetic framework (1,6). We used MLST and three other conventional epidemiologicaltyping approaches to describe the genetic population structureof strains that were systematically collected from nonhospitalizedindividuals in the community. This way we were able to objectivelyassess the value of typing techniques deployed for epidemiologicalpurposes.

MATERIALS AND METHODS

Top

Materials and Methods

Strains of S. aureus carried nasally were isolated from a representativesample of nonhospitalized elderly individuals ( $>=$ 65 years old)living in the Greater Nottingham Health District (Nottinghamshire,United Kingdom) (3). A random sample (117 of 257 of this straincollection) was subjected to four different typing techniquesthat included MLST, pulsed-field gel electrophoresis (PFGE)(SmaI macrorestriction analysis), randomly amplified DNA analysis(RAPD), and phage typing. All typing methods were performedby standardized published protocols (1, 9, 11), and genotypeswere classified according to conventional criteria (1, 4, 7,10). A standardized protocol for PFGE can be found at the PublicHealth Laboratory Service website (http://www.phls.org.uk/inter/harmony/New%20PFGE%20protocol.pdf).Typing results were stored, processed, and analyzed using thestatistical suite of BioNumerics software (Applied-Maths, Ghent,Belgium). An index of diversity, defined as the probabilitywith which two isolates chosen at random will be of a differenttype, was used to measure the frequency with which organismsof a particular type occur in the population of strains carriedand serves at the same time as an unbiased estimate of the discriminatoryability of the relevant typing method (5, 8). Confidence intervalsfor discriminatory indices were calculated as previously described(2).

RESULTS

Top

Results

MLST. MLST identified 46 different allelic profiles or sequence types(STs) among 117 strains carried by individuals in the generalcommunity. Sixteen STs were represented by more than one isolate.The genetic diversity as determined by MLST was 95.7% (95% confidenceinterval, 94.4 to 97.1%). We used Based Upon Related SequenceTypes (BURST) analysis as a comprehensive method to classifydifferent STs that demonstrate a recent phylogenetic relationshipas described at the Multi Locus Sequence Typing website (www.mlst.net/BURST/BURSTREADME.htm).The classification was based on the data for 807 isolates ofS. aureus for which MLST types were available by October 2001.Of the 117 total isolates, 105 could be assigned to a clonalcomplex, and the sample comprised all of the 13 clonal complexesidentified so far. However, three clonal complexes were representedonly by single isolates in our collection, and genotypes thatdid not correspond to any clonal complex were defined as singletons.

PFGE. SmaI macrorestriction analysis identified 57 PFGE types whenthree and more band differences indicated a different type.The ability of this method to discriminate strains carried byindividuals in the community was 97.6% (96.8 to 98.5%) and wasbetter but not significantly different from MLST (Table 1).Importantly, there was a high degree of concordance betweenthe results by PFGE and MLST. Most isolates (98 of 105) assignedto clonal complexes by MLST and BURST analyses also clusteredin the PFGE dendrogram after unweighted paired group mean analysis(UPGMA) clustering using band-based similarity values (Fig.1), whereby PFGE patterns differed by a maximum of five bandswithin clonal complexes.

View this table:
[in this window]
[in a new window]
TABLE 1. Comparison of typing results for 117 S. aureus strains carried by individuals in the community^a

View larger version (23K):
[in this window]
[in a new window]
FIG.1. UPGMA dendrogram illustrating similarity (based on Dice coefficients) of macrorestriction profiles (SmaI) (PFGE) of 117 isolates of S. aureus carried by individuals in the community. The scale at the top shows percent similarity. Abbreviations: CC, clonal cluster as defined by MLST and BURST analyses; ST, sequence type (MLST); #, isolate number; sg, singleton not grouped in a clonal cluster.

RAPD. DNA fingerprints generated with a single arbitrary primer identified16 different RAPD types, resulting in a discriminatory capacityof 85.7% (82.9 to 88.6%). Thus, the ability to discriminatedifferent strains by RAPD when using only one primer remainedsignificantly restricted compared to those of PFGE and MLST.Still, there was a fair degree of agreement between the pairwisesimilarity values generated by MLST and RAPD, as indicated bythe high overall correlation value.

Phage typing. Of the 117 isolates, 28 were not typeable by the standard methodwith 23 phages of the Basic International Set at routine testdilution (typeability of 76.1%). The remaining isolates weregrouped into 23 phage types corresponding to a discriminatoryability of 80.5% (75.1 to 85.8%). Notably, the overall correlationwith the other typing techniques was low, indicating a poorconcordance between phage typing and DNA-based approaches.

Group violation. Comparing all pairwise within group similarity values with betweengroup similarity values allows an objective assessment of therobustness of the grouping of typing techniques. Within groupsimilarity values equal to or less than those of between groupvalues violate the group assignment. We calculated the percentagegroup violation as the proportion of pairwise similarity valuesthat did not allow an unequivocal assignment to the clonal complexesdefined by MLST and BURST analyses. We were thus able to determineto what extent results of typing techniques other than MLSTare able to predict clonal complexes. PFGE allowed the bestprediction, with 78.4% of all similarity values not in conflictwith the classification; results for RAPD were fair (50.6%),whereas only 13.1% of phage typing data allowed an unambiguousprediction of the clonal complex (Table 1).

DISCUSSION

Top

Discussion

We chose a computer-based algorithm that defines clonal complexesas groups in which each isolate is identical to at least oneother isolate at five or more of the seven loci identified byMLST. This approach arrives at a rational classification byapplying simple phylogenetic rules. Conversely, band-based typingtechniques, such as macrorestriction analysis or RAPD, facedifficulties when unequivocal types need to be assigned, asgradual changes in the positions of otherwise anonymous bandsmake a designation arbitrary. The genetic population structureof strains carried by individuals showed a predominance of afew clonal complexes. Moreover, the genetic diversity did notdiffer significantly from a sample of hospital- and community-acquireddisease isolates recently collected in Oxfordshire (N. P. Day,personal communication). There is thus no evidence from MLSTcontradicting the assumption that isolates from cases of invasivedisease represent a random sample of strains from the reservoirof S. aureus strains carried nasally in humans. By addressingthe frequencies of naturally occurring strains carried by humans,we were also able to determine the positive predictive valuewith which each typing method can identify an epidemiologicalassociation (such as person-to-person transmission) when a pairof indistinguishable strains is isolated from two distinct carriers,and one would consider the lower confidence value of the discriminatoryability as a conservative estimate to this effect. We regardthis as extremely valuable information when interpreting typingresults for clinical practitioners or an increasingly informedpublic. Clearly, MLST and PFGE are the most powerful techniquesfor high-precision epidemiological typing when episodes of person-to-persontransmission should be ascertained. RAPD and phage typing, bothrapid and able to process large amounts of strains, seem, however,unsuitable for this purpose. Most importantly, MLST providesan adequate tool for producing genetic profiles for a vast numberof isolates especially in nonepidemic circumstances, i.e., fornational reference services or when comparing large internationalstrain collections.

We found that DNA-based typing schemes come to very similarconclusions and combining different techniques will not increasethe discrimination of strains of S. aureus to a large extent.Moreover, it appears that PFGE patterns could theoreticallypredict clonal complexes, pointing to the presence of a largelycongruent phylogenetic signal in different regions of the staphylococcalgenome. Future work should identify the most informative nucleotidesequences that could form the basis of a versatile strain designationof S. aureus.

FOOTNOTES

* Corresponding author. Mailing address: Division of Microbiology and Infectious Diseases, Queen's Medical Centre, University Hospital Nottingham, Nottingham NG7 2UH, United Kingdom. Phone: 44-115-970 9162. Fax: 44-115-970 9233. E-mail: Hajo.Grundmann{at}Nottingham.ac.uk.

REFERENCES

Top