Campylobacter jejuni in Different Canine Populations: Characteristics and Zoonotic Potential

Abstract

With most epidemiological studies focused on poultry, dogs are often overlooked as a reservoir of Campylobacter, even though these animals maintain close daily contact with humans. The present study aimed to obtain a first insight into the presence and characteristics of Campylobacter spp. in different canine populations in Portugal, and to evaluate its zoonotic potential through genomic analysis. From a total of 125 rectal swabs collected from companion (n = 71) and hunting dogs (n = 54) living in two different settings, rural (n = 75) and urban (n = 50), 32 Campylobacter spp. isolates were obtained. Four different Campylobacter species were identified by Multiplex PCR and MALDI-TOF mass spectrometry, of which Campylobacter jejuni (n = 14, 44%) was overall the most frequently found species. Relevant resistance phenotypes were detected in C. jejuni, with 93% of the isolates being resistant to ciprofloxacin, 64% to tetracycline, and 57% to ampicillin, and three isolates being multi-drug-resistant. Comparison of the phenotypic and genotypic traits with human isolates from Portuguese patients revealed great similarity between both groups. Particularly relevant, the wgMLST analysis allowed the identification of isolates from human and dogs without any apparent epidemiological relationship, sharing high genetic proximity. Notwithstanding the limited sample size, considering the high genomic diversity of C. jejuni, the genetic overlap between human and dog strains observed in this study confirmed that the occurrence of this species in dogs is of public health concern, reinforcing the call for a One Health approach.

Keywords: Campylobacter spp.; One Health; dogs; whole genome sequencing.

Figure 1

ERIC-PCR fingerprints of Campylobacter jejuni isolated in dogs. The ID of the sample is presented in white. The letter and numbers in yellow represent the cluster assigned to the profile.

Figure 2

ERIC-PCR fingerprints of Campylobacter jejuni ST-22 and ST-6461 (presented in yellow) in a 2% agarose gel. Lane M, 1 Kbp Plus ladder (Invitrogen, Waltham, MA, USA). The ID of the sample is presented in white in the case of the dog isolates and in blue in the human isolates. Finally, a comparative gene-by-gene analysis was performed between PT isolates and other ST-6461 (n = 91) and ST-22 (n = 83) with available genomes from different regions and sources (Supplementary Tables S1 and S2), with the results represented by minimum spanning trees (MSTs) (Figure 3 and Figure 4).

Figure 3

Genomic relatedness among Campylobacter jejuni ST-6461 strains based on a dynamic gene-by-gene approach using a wgMLST schema with 2795 loci. (A) The Minimum spanning tree was constructed using the goeBURST algorithm implemented in the PHYLOViZ online platform and is based on the allelic diversity found among the 1011 genes shared by 100% of the validated strains. Filled circles (nodes) represent unique allelic profiles, and are colored according to strains’ source (human, chicken, or dog). The size of nodes is proportional to the number of isolates it represents. Portuguese strains are marked with a green circle: three isolates from dogs (PT_Cj-Br2/1, PT_Cj-Br7/2, PT_Cj-Br8/1) and one from human (PT_Cj-4010). The numbers in grey on the connecting lines represent the allele differences (AD) between strains. For better visualization, only AD ≥ 5 are shown. (B) Box plot depicting the AD (including the median values) within and between PT strains from dogs and the remainder.

Figure 4

Phylogeny of Campylobacter jejuni ST-22 strains based on a dynamic gene-by-gene approach using a wgMLST schema with 2477 loci. (A) The initial MST was constructed based on the allelic diversity found among the 924 genes shared by 100% of the validated strains. Portuguese strains are marked with green circles: four from dogs (A11, Br8/2, Br4/2, Br9) and four from humans (3977, 4020, 4093, 4113); US strains are marked with purple outer rings; the remaining strains are from the UK. (B) Sub-MST reconstruction based on the allelic diversity found among the 953 genes shared by all validated PT strains (ID strains in bold). For both panels, trees were constructed using the goeBURST algorithm implemented in the PHYLOViZ Online platform. Filled circles (nodes) representing unique allelic profiles are colored according to strains’ source (human, chicken or dog). The size of nodes is proportional to the number of isolates it represents. The numbers in grey on the connecting lines represent the allele differences (AD) between strains. For better visualization, only AD ≥ 5 are shown.