Molecular analysis of human and canine Staphylococcus aureus strains reveals distinct extended-host-spectrum genotypes independent of their methicillin resistance

Abstract

Staphylococcus aureus causes a wide range of infectious diseases in humans and various animal species. Although presumptive host-specific factors have been reported, certain genetic lineages seem to lack specific host tropism, infecting a broad range of hosts. Such Extended-Host-Spectrum Genotypes (EHSGs) have been described in canine infections, caused by common regional human methicillin-resistant S. aureus (MRSA) lineages. However, information is scarce about the occurrence of methicillin-susceptible S. aureus (MSSA) EHSGs. To gain deeper insight into EHSG MSSA and EHSG MRSA of human and canine origin, a comparative molecular study was carried out, including a convenience sample of 120 current S. aureus (70 MRSA and 50 MSSA) isolates obtained from infected dogs. spa typing revealed 48 different spa types belonging to 16 different multilocus sequence typing clonal complexes (MLST-CCs). Based on these results, we further compared a subset of canine (n = 48) and human (n = 14) strains, including isolates of clonal complexes CC5, CC22, CC8, CC398, CC15, CC45, and CC30 by macrorestriction (pulsed-field gel electrophoresis [PFGE]) and DNA-microarray analysis. None of the methods employed was able to differentiate between clusters of human and canine strains independently of their methicillin resistance. In contrast, DNA-microarray analysis revealed 79% of the 48 canine isolates as carriers of the bacteriophage-encoded human-specific immune evasion cluster (IEC). In conclusion, the high degree of similarity between human and canine S. aureus strains regardless of whether they are MRSA or MSSA envisions the existence of common genetic traits that enable these strains as EHSGs, challenging the concept of resistance-driven spillover of MRSA.

Fig 1

Macrorestriction analysis of representative canine and human isolates. SmaI macrorestriction patterns are shown with a dendrogram (cluster analysis; Dice coefficient [1.2% tolerance and 0.5% optimization]) displaying 43 canine and 12 human isolates with spa type, PFGE type, and host origin. Strains are grouped into clonal complexes according to the spa type and macrorestriction pattern. Abbreviations: CC, clonal complex; BW, Baden Wuerttemberg, Germany; MP, Mecklenburg Pomerania, Germany; NRW, North Rhine Westphalia, Germany; RP, Rhineland Palatinate, Germany; LS, Lower Saxony, Germany; resp. syst., respiratory system; n.t., not typed; pos, positive; neg, negative.

Fig 2

Microarray hybridization results for agr type and capsule type and a selection of resistance- and virulence-associated genes. Black squares, positive; gray squares, ambiguous results according to Alere DNA-microarray analysis software; divided squares, variable results for individual isolates within grouped strains per lane. The eight canine strains indicated by the superscript “a” among the CC22 MRSA strains are VB 994453.1, VB 980511, VB 976714.1, IMT 8729, VB 969045, IMT 9715, VB 953142, and VB 986856.2.

Fig 3

Rendered tree based on the microarray hybridization results of 334 target sequences after cluster analysis of the data performed using Pearson correlation and UPGMA. Canine isolates are black; human strains are highlighted in green. The branch lengths display the percentages of different loci in comparison to other isolates. Each strain is shown with its geographic origin. B, Berlin, Germany; BV, Bavaria, Germany; BW, Baden Wuerttemberg, Germany; H, Hesse, Germany; LS, Lower Saxony, Germany; MP, Mecklenburg Pomerania, Germany; NRW, North Rhine Westphalia, Germany; RP, Rhineland Palatinate, Germany; SAX, Saxony, Germany; AUT, Austria; FRA, France; ITA, Italy; NEL, Netherlands; NOR, Norway; SWE, Sweden.