Egg genotyping reveals the possibility of patent Ancylostoma caninum infection in human intestine

Hookworms are intestinal parasites that cause major public health problems, especially in developing countries. To differentiate eggs from different hookworm species, it is necessary to use molecular methodologies, since the eggs are morphologically similar. Here, we performed the molecular identification of single hookworm eggs from six Brazilian states. Of the 634 eggs individually analyzed, 98.1% (622/634) represented Necator americanus, and surprisingly, 1.9% (12/634 eggs from the same patient) represented Ancylostoma caninum. DNA analysis of the A. caninum-positive stool sample revealed no contamination with animal feces. This is the first report of the presence of A. caninum eggs in human feces, which may have a direct implication for the epidemiology of hookworm infection caused by this species. This suggests the need for special attention regarding prophylaxis, as different reservoirs, previously not described, may have great relevance for the spread of A. caninum.

www.nature.com/scientificreports www.nature.com/scientificreports/ fragment length polymorphism (RFLP-PCR) 12,20 , high resolution melting (HRM) 21 , quantitative real-time multiplex PCR 22 , single-strand conformation polymorphism (SSCP) 23 and sequencing 24,25 . All of these studies performed analyses of adult worms and egg/larvae pools. We performed our analyses by conventional PCR by using a method previously standardized by Sahimin and coworkers 16 , with some modifications. To the best of our knowledge, this is the first study to identify hookworm species from single eggs. This is important because individual analysis allows the accurate detection of species, whereas, depending on the sensitivity of the technique, egg/larvae pool analysis can detect only the most prevalent species in the sample. In addition, obtaining adult worms for morphological analysis is unfeasible for obvious reasons.
Of the 634 eggs evaluated, 98.1% (622/634) were identified as N. americanus. In fact, according to Loukas and coworkers 5 , N. americanus is the predominant human hookworm, which globally accounts for the majority of patent human hookworm cases and is especially common in the Americas. These data corroborate the molecular analyses of Monteiro and coworkers 25 , who evaluated hookworm populations in humans in northeastern Brazil. Brooker and coworkers 26 , by using morphological analysis, observed that all adult hookworms recovered from populations in southeastern Brazil were N. americanus, while Marzochi and Chieffi 27 , by using the morphological analysis of larvae obtained after fecal culture, observed the presence of A. duodenale in 14.2% of the analyzed samples. Studies in other South American countries, such as Ecuador, have also described A. duodenale in the intestines of humans 28 . In the present study, no A. duodenale eggs were detected. Biological characteristics linked to the development of N. americanus explain why this species is the most prevalent. The eggs of this species do not develop at temperatures below 15 °C, and the larvae prefer shady, moist areas with temperatures at or above 30 °C 29 .
Ancylostoma ceylanicum, primarily described as a dog and cat nematode, is highly prevalent in humans in Southeast Asia, including Malaysia and Indonesia 30 . Although it has been consistently used as an experimental model for studies in Brazil 31,32 , A. ceylanicum has never been described in animal or human populations in this country. However, the migratory flow of tourists and workers to Brazil observed in recent decades must be carefully analyzed, since this migration has already been responsible for introducing A. ceylanicum species to countries in which it had not been previously described 33 .
Although many studies conducted worldwide have revealed a high prevalence of A. caninum in canids and felids 13,34 , the infection by this species of humans has so far been restricted to skin infections that have mainly been acquired in parks and coastal regions 35 . Nevertheless, our results reveal for the first time the presence of A. caninum eggs in human feces. This may have a direct implication for the epidemiology of human parasitism by this species, since the source of infection could not be restricted to infected animals only. This should be critically studied further, especially since control strategies have been focused on classically described hosts (dogs and cats) in terms of vaccine development and coverage. Indeed, in recent decades, the target glutathione-s-transferase-1 has been studied as a vaccine antigen for N. americanus (Na-GST-1) 36 and A. caninum (Ac-GST-1) 37 . Although this vaccine has resulted in a decreased number of adult worms and fecal eggs in dogs 37 , tests of the efficacy of this vaccine in humans have never been performed 36 . Because vaccine targets are species-specific, variation in hookworm species in a geographical region may negatively influence vaccine efficacy. In fact, regional variability may even impact anthelmintic efficacy; mutations involved in drug resistance in A. caninum collected in the field from different regions of Brazil have been described by our group 38 .
Although A. caninum eggs have never been observed in human feces, George and coworkers 11 detected DNA from A. caninum in a fecal sample. These studies did not allow us to conclude that the eggs present in the samples were from this species, since the analyses were made based on DNA extraction from feces that also had N. americanus (which has eggs similar to those of A. caninum). According to previous authors, immature forms in the intestines of humans could release DNA into the human intestine, which would be detected by molecular methods. Our analyses, having been performed using DNA from a single egg, allow us to state with confidence that 1.9% (12/634) of the eggs represented Ancylostoma spp., which were identified as A. caninum by sequencing.
One hypothesis that could be proposed to explain the presence of A. caninum eggs in human feces would be the ingestion of eggs present in contaminated water and food and the consequent intact excretion of these eggs in feces. This has already been demonstrated in dogs with coprophagic habits, which can excrete Eimeria spp. and non-dog typical helminth eggs 39,40 . This is especially common for plant nematodes that are commonly ingested during human feeding, such as a parasite of the genus Meloidogyne 41 , causing infected patients to present with eggs in their feces without completion of the parasite cycle. However, hookworm eggs have a single and very thin membrane that would most likely be destroyed during passage through the stomach of humans 42 . Thus, the morphological characteristics of A. caninum eggs make this hypothesis strongly untenable. In addition, by observing the stage of egg development under optical microscopy, there was an absence of embryonated eggs in the feces. In case of the eggs simply being in transit through the gastrointestinal tract, it would seem probable that they would already be embryonated (see Supplementary Fig. S1).
A. caninum eggs found in a single patient sample could suggest contamination with fecal material from dogs and cats. However, DNA extraction from stools followed by molecular analyses to identify possible contaminants, performed in duplicate, revealed that only human DNA was found in the sample. This suggests that the eggs found did not come from the feces of animals infected with A. caninum. It has also to be considered that the potential uptake of animals A. caninum eggs does not necessarily mean that relevant amounts of animals faeces need to be taken up, since eggs may have been attached to food. In fact, many studies report the presence of hookworm eggs in food 43,44 and also in water 45 . However, this a very low probability since twelve A. caninum eggs were found in the respective patient. In addition, some wildlife may be considered A. caninum hosts. Our analyzes from the feces of the patient with A. caninum would not be able to detect DNA from some of these animals, such as coyotes 46 . However, contamination of samples with material from these animals seems unlikely.
Unfortunately, we do not have any information regarding the patients, that include data related to the lifestyle habits of individuals, such as domesticated animals and work activities. Since dogs and cats are considered the main definitive hosts of A. caninum, living with infected animals could be a risk factor for the acquisition of the parasite 20 , either by causing larva migrans infections or by its development in the human intestine. In addition, work activities such as farm workers and other services that require contact with the soil can be considered a risk factor 34 . It is also important to consider the possibility of human host immunosuppression, since the absence of an effective immune response may have allowed the development and maturation of the nematode in the individual's intestine.
Humans and canines have shared the same evolutionary factors for many centuries, so the domestication of dogs led to the sharing of microbiota, pathogens and even similar eating habits 47 . This may have consequences on parasite transmission and reservoirs, such as Strongyloides, Toxocara, tapeworms and hookworms that can be passed mutually from dogs and humans 48 . Considering that population genetic studies denote a very high phylogenetic proximity between A. caninum, N. americanus and A. duodenale 49 , it seems reasonable that evolutionarily A. caninum has acquired the possibility of becoming sexually mature adult in human intestines. In fact, A. ceylanicum is originally a canine and feline parasite, but which over the years has acquired the ability to infect humans 11 .
Our results show the possibility of infection and reproduction of A. caninum in the human intestine. We suggest that additional analyses should be performed to determine the infection pathways. According to Landmann and Prociv 10 , patients with eosinophilic enteritis caused by A. caninum acquired the infection percutaneously. In fact, migration of the hookworm larvae into skeletal muscles has also been reported in an individual with a large cutaneous exposure, with recovery of third-stage larvae within a muscle fiber 50 . Normally, infection with this nematode occurs by the ingestion of filarioid larvae or penetration of these larvae into the skin of animals 3,4 . To complete their development, larvae need to be able to produce various enzymes involved in connective tissue degradation, such as hyaluronidases, which are often species specific 5,51,52 . Could this indicate the adaptation of the larvae of this species, which in recent times have been able to infect humans? Considering the proximity of the classical hosts to the human species, has A. caninum acquired the ability to develop and reproduce in the intestines of humans, or has this just not been reported so far? These are extremely important questions that may lead to new chapters in the history of parasitology. Methods ethical statements. This work was approved by the Comitê de Ética em Pesquisa -COEP (CAAE 61047216.7.0000.5149) at the Universidade Federal de Minas Gerais (UFMG). As we used human feces obtained from commercial laboratories performing pathological analysis, an informed consent document was not required. We did not obtain any subject identification, and the data were analyzed anonymously.
Sampling and DnA extraction. The stool samples were processed, and DNA extractions were performed for single hookworm eggs from patients collected in six Brazilian states exactly as described previously by Zuccherato and coworkers 53 . The initial isolation of the eggs was performed according to Ritchie 54 with modifications, and samples that were positive for hookworm were stored in 10% formaldehyde for later molecular analysis. In summary, 2 ml of stool suspension was homogenized, filtered through gauze and transferred to a 15 ml tube. Five milliliters of sulfuric ether were added to the suspension and then stirred vigorously, followed by 1 min of centrifugation at 14,000 × g. The supernatant was discarded. Eggs were washed in an additional step by adding 500 μl of 1% hypochlorite for 10 min to the samples. The material was centrifuged at 14,000 × g, and the supernatant was discarded. The eggs were washed again using 500 μl of ultrapure water, followed by centrifugation at 14,000 × g. The supernatant was then discarded. The pellet was resuspended in 100 μl of ultrapure water. For DNA extraction, the eggs were observed under an optical microscope, individually pipetted into a volume of 1 μl and transferred to a 500 μl microcentrifuge tube containing 10 μl of buffer, as described by Lake and coworkers 55 . The eggs were incubated for 2 h at 57 °C, followed by incubation for 20 min at 80 °C for proteinase K inactivation. The material was stored at −20 °C until use. In total, 634 hookworm eggs from 53 patients collected in six Brazilian states were analyzed. Table 1 shows the collection sites, the number of patients and the number of eggs collected from each state. Differentiation of eggs from N. americanus and Ancylostoma spp. The egg analyses for species determination were performed according to Sahimin and coworkers 16 , with some modifications. After DNA extraction from single eggs, conventional PCR was performed with three primers in the same reaction (forward www.nature.com/scientificreports www.nature.com/scientificreports/ NA and AD1 and reverse NC2). Table 2 shows the sequences of each primer used in this study and the respective reference. These primers were used to amplify a region in the internally transcribed spacer 2 and 28 S ribosomal RNA (ITS2-28S rRNA). The primers NA and AD1 will only anneal to DNA from N. americanus and Ancylostoma spp., respectively, while primer NC2 will anneal to DNA from many hookworm species; the NA + NC2 combination would result in a 250 bp fragment for N. americanus and no amplification for Ancylostoma spp., while the AD1 + NC2 combination would result in a 130 bp fragment for Ancylostoma spp. and no amplification for N. americanus.

Number of patients
All PCR amplifications were performed in a 10 μl reaction containing the three primers (NA, AD1 and NC2) (0.2 μM each), Taq DNA polymerase (1 U) (Phoneutria, Brazil), deoxynucleoside triphosphate (dNTPs) (200 μM each), reaction buffer (1X), 5 μl of single egg DNA (not previously quantified to prevent material loss) and ultrapure water. The PCR amplifications were performed according to the following program: 94 °C for 5 min, followed by 35 cycles of 94 °C for 1 min, 55 °C for 1 min and 72 °C for 1 min, and a final extension at 72 °C for 7 min. A "blank" sample was included in all amplification runs in which the DNA was replaced with water to assess the presence of possible contaminants. The reaction product was subjected to electrophoresis in a 1% agarose gel (w/v) (Midsci, USA) in 0.5X TAE buffer, and the gel was stained with GelRed ™ (Biotium, USA). controls. DNA from N. americanus and Ancylostoma spp. were previously extracted from adult worms 55,56 .
PCRs were performed under the same conditions as previously described above with 40 ng of DNA from each parasite and the primer combinations NA + NC2 (for N. americanus) and AD1 + NC2 (for Ancylostoma spp.) (ITS2-28S rRNA). The products were then purified (Illustra GFX PCR DNA and Gel Band Purification Kit, GE Healthcare, UK), and the DNA concentration was determined. The controls fragments were subsequently cloned using the pGEM-T Easy Vector System (Promega, USA), transformed into XL1-blue cells (Phoneutria, Brazil) and recovered via minipreps (Wizard Plus Miniprep DNA Purification System, Promega, USA). The plasmids were sequenced, and the species were successfully identified. A total of 20 ng of plasmid was used as a control for the reactions for the N. americanus and Ancylostoma spp.
Species differentiation of Ancylostoma spp. The molecular technique adopted in this study allowed the differentiation between the species N. americanus and the genus Ancylostoma, but for the differentiation between possible Ancylostoma spp. species, sequencing reactions were performed using the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, USA) with a ABI 3130 × 1/Genetic Analyzer automated sequencer (Applied Biosystems, USA). These samples were not cloned, but were sequenced in both, forward and reverse directions, and chromatogram analysis was performed using FinchTV software (Geospiza, USA).
Determination of possible contaminants. The fecal sample that contained A. caninum eggs underwent a process of DNA extraction to determine the presence of possible contamination with feces from these animals. For this, DNA extraction was performed from 220 mg of stool with the QIAamp ® DNA Stool kit (Qiagen, Germany) according to the manufacturer's recommendations. The DNA was quantified, and a total of 40 ng was used in three PCRs with different primer pairs. The Camt1F + Camt1R (167 bp), Dmt1F + Dmt1R (193 bp) and Hmt2F + Hmt2R (162 bp) primer combinations were designed to anneal only to DNA from the Felis catus, Canis familiaris and Homo sapiens species, respectively. These primers were used to amplify a region of the Cytochrome c oxidase subunit I (COI). Table 2 shows the sequences of each primer used and the respective reference. PCRs were performed under the same conditions as previously described above, and the sequencing of the amplified material was performed. Previously extracted DNA from F. catus, C. familiaris and H. sapiens were used as reaction controls.  Table 2. Primers used for the molecular determination of hookworm and host species.