Fecal samples collected from the rectal ampulla of 220 dogs were preserved in 70% ethanol for molecular analysis. We isolated genomic DNA from the samples (≈200 mg each) by using the QIAmp PowerFecal Pro DNA Kit (QIAGEN, https://www.qiagen.com), eluted into 50 μL (instead of 100 μL) of solution C6, and then stored the samples at –20°C. To differentiate the hookworm species, we subjected individual DNA samples to M-qPCR (3) and analyzed and displayed the data by using GraphPad Prism version 8.0 (GraphPad, https://www.graphpad.com). Samples positive for A. ceylanicum hookworm DNA by quantitative PCR underwent conventional PCR and Sanger sequencing targeting a partial region of the cytochrome oxidase subunit 1 (cox1) gene as previously described (8). To analyze sequences, we used Geneious prime 2021, version 2.2 (https://www.geneious.com).

Figure

Figure. Proportion of dogs infected with zoonotic hookworms in Grenada, 2021, determined by using multiplex quantitative real-time PCR. Error bars indicate 95% CI. A., Ancylostoma; U...

Overall, of the 211 samples for which DNA isolation was successful, 40.8% (95% CI 34.1%–47.3%) were positive for >1 hookworm species: 27.5% (95% CI 21.4%–33.5%) A. caninum, 4.7% (95% CI 1.87%–7.61%) A. ceylanicum, and 8.5% (95% CI 4.76%–12.3%) both parasites. Neither A. braziliense nor Uncinaria stenocephala DNA was detected (Figure). BLAST (https://blast.ncbi.nlm.nih.gov) analyses of cox1 sequences showed 100% nt identity with sequences of A. ceylanicum hookworms isolated from humans (GenBank accession nos. MK7928230-35) and dogs (GenBank accession no. LC533320). A representative cox1 sequence of A. ceylanicum hookworms found in this study was deposited in GenBank under accession no. OP077312.

Molecular analysis identified A. ceylanicum hookworms in dogs in the Americas. In the past decade, unexpectedly high prevalence and distribution of A. ceylanicum hookworms in dogs, wildlife, and humans has been reported mainly in the Asia–Pacific region but also in Africa (1,2,9). A. ceylanicum was erroneously considered synonymous with A. braziliense, until the 2 hookworm species were differentiated on the basis of configuration of the lateral bursal rays of the male (1). Hence, the “emergence” of A. ceylanicum hookworms may be associated with the increased use of molecular tools to specifically identify and differentiate these parasites in areas where multiple species of hookworm co-exist (1,2).

In the Americas, A. ceylanicum hookworms have not been considered etiologic agents of human infection, despite isolation of adult worms in 4 of 64 human cadavers in the Amazonas, Brazil, in the early 1920s and, more recently, reported detection of this parasite in travelers returning from countries in Latin America (e.g., Colombia and French Guiana) to France (1,5,10). Of note, the geographic distribution of A. ceylanicum hookworms in dogs strongly mirrors that in humans in the Asia–Pacific region (2). To date, only a few coproscopic studies of hookworms in dogs in Grenada have been undertaken (11), but absence of molecular investigation raises questions as to the specific identity of the hookworms on the island.

In humans, other species of hookworms of canids can cause hookworm-related cutaneous larva migrans. A. braziliense hookworms cause severe serpiginous “creeping eruptions,” and A. caninum hookworms cause peripheral eosinophilia and aphthous ileitis (1). Although recent cases of long-lasting hookworm-related cutaneous larva migrans have been reported in the Caribbean, including Grenada (11,12), Martinique, Jamaica, and Brazil (1), we did not detect A. braziliense hookworm infection in our canine cohort. Nonetheless, cases of hookworm-related cutaneous larva migrans in some countries in Latin America have been attributed to hookworms of canids, despite lack of molecular evidence. For instance, 4 cases were detected in 1983 during a hookworm outbreak among US military soldiers returning from Grenada, but the etiology of these infections was not established (11). Furthermore, although some circumstantial evidence indicates that A. caninum hookworms cause patent infections in humans (9,13), molecular evidence is lacking, which is particularly pertinent given the close genetic relatedness of A. caninum and A. duodenale hookworms (1,13). Thus, use of molecular methods to specifically detect, identify, and differentiate is crucial for surveillance of hookworm infections/disease in dogs and humans and for monitoring the success of control campaigns in hookworm-endemic areas (14).

Molecular detection of A. ceylanicum and A. caninum and the absence of A. braziliense hookworms in dogs in Grenada demonstrates the value of using molecular techniques to accurately identify hookworm species in dogs and humans living in the Americas. The climate in Grenada is similar to that in other countries where prevalence of zoonotic hookworms is high (1,6,15). According to the World Bank in Grenada, 34.3% of its population still live in households considered multidimensionally poor, where solid waste management and public sewer systems remain inadequate (6). Those factors, together with lack of appropriate veterinary care for dogs and closeness of human–dog interactions, favor transmission of zoonotic pathogens to humans. Given the significance of A. ceylanicum hookworms as agents of human hookworm infection in the Asia–Pacific region (1,2), we advocate for active surveillance of this zoonotic parasite in dogs and humans living in tropical and subtropical countries in the Americas by using reliable molecular diagnostic tools targeting appropriate genetic markers.