Evidence of Helicobacter spp. in Saliva and Gastric Mucosa of Domestic Dogs in the Central Region of Rio Grande do Sul, Brazil

Helicobacter pylori is a spiral-shaped bacterium, which plays a role in the aetiology of gastric diseases in humans. Non-H. pylori Helicobacter (NHPH) species naturally colonise the stomach of animals and also induce gastric lesions in humans, highlighting their zoonotic importance. We evaluated the gastric bacterial colonisation density and gastric lesions and sought to identify the main phylogenetic groups of the Helicobacter spp. obtained from dogs in the central region of Rio Grande do Sul, Brazil, with this study aiming to investigate the occurrence of Helicobacter spp. in saliva and gastric samples from these dogs. This study included 35 dogs and used analysis such as cytology, histopathology, PCR, rapid urease testing, and phylogenetic analysis. Of the dogs, 94.3% were positive for Helicobacter spp., and these bacteria were present in the stomach of 32 dogs and saliva of eight. Respectively, eight, 15, and nine dogs had mild, moderate, and severe colonisation. Lymphocytic-plasmacytic infiltrate was the main gastric lesion. However, the presence of Helicobacter and the density appeared to be unrelated to the gastric lesions. The samples possessed a high nucleotide identity with remarkably similar sequences among some of the species of NHPH such as H. heilmannii s.s., H. salomonis, H. felis, and H. bizzozeronii. The saliva of domestic dogs, even of those who appear clinically healthy, can cause Helicobacter infection in humans and other animals, with, in these dogs, increased density, occurrence rate, and predominance of NHPH of zoonotic importance being found in the stomach with a lower occurrence of Helicobacter spp. in the saliva.


Introduction
Helicobacter spp. are spiral-shaped mobile Gram-negative bacteria with tropism for the gastric mucosa of humans and animals [1]. Helicobacter pylori (H. pylori) was the first species isolated from the Helicobacter genus by Australian researchers Marshall and Warren in 1983 [2]. Further research has shown that humans are the natural host of H. pylori and have established this species as the primary aetiology of peptic ulcers and gastric neoplasms [3]. Years later, in samples of gastric mucosa from humans, another spiral-shaped bacteria with similar morphological features to H. pylori was documented, and this microbe was found to be remarkably similar to the spiral-shaped bacteria found in the gastric mucosa of domestic animals [4]. is new spiral bacteria was later classified by Solnick et al. [5] as Helicobacter heilmannii. However, phylogenetic studies were later employed utilising the 16 rRNA target gene, and these studies found instead that this bacteria belonged to a range of several Helicobacter species isolated from domestic and wild animals, such as H. felis, H. bizzozeronii, H. salomonis, Helicobacter heilmannii sensu stricto (H. heilmannii s.s.), H. bílis, H. cynogastricus, and H. baculiformis [6]. erefore, in order to organise this large group of bacteria into a single term, they were consensually denominated as non-Helicobacter pylori Helicobacter (NHPH) [7].
NHPH are the target of various studies due to their relationship with upper digestive tract illnesses in humans and their zoonotic importance [8]. Dogs are the natural hosts of NHPH and harbour this bacteria in their gastric mucosa, gut, and oral cavity; thus, gastric juice, saliva, and faeces are possible sources of transmission for this bacteria to infect humans [9][10][11]. In dogs, the main species of NHPH found are H. heilmannii s.s, H. bizzozeronii, H. salomonis, H. felis, and H. canis [9,11].
Chronic inflammation of the gastric mucosal tissue, peptic ulcers, and gastric mucosa associated lymphoid tissue lymphoma are the clinical alterations described in humans with NHPH infection [12,13]. In human populations, NHPH has a prevalence of 0.5% in developed countries [14] and 6.2%-15% in underdeveloped countries [15,16]. us, countries with lower socioeconomic development tend to have a higher prevalence of infected people with NHPH [17]. However, information regarding the importance of domestics dogs as reservoirs for this bacteria and the data related to the number of NHPH occurrences in the canine population of these countries has not yet been explicated [18].
Due to the zoonotic implications of NHPH, compounded by the high density of domestic animals, and sanitary problems affecting Brazil, it is important to elucidate this information to assist with future studies of public health. Moreover, the geographic variation could affect the prevalence of the Helicobacter species [11,18]. us, the aim of this study was to investigate the presence of Helicobacter spp. from saliva and the gastric mucosa of domestic dogs. Moreover, this study documented the gastric bacterial colonisation density, the number of gastric lesions present and sought to identify the main phylogenetic groups of Helicobacter spp. found in dogs from the central region of Rio Grande do Sul.

Ethics Statement.
All the procedures were performed with the consent of owners of the dogs and the study protocol was approved by the Animal Ethics Committee of Universidade Federal de Santa Maria (Approval number 2827081018) and conducted in accordance with national guidelines and regulations for the care and use of laboratory animals established by the National Council for Animal Experimentation Control (CONCEA), Brazil.

Animals.
is study collected saliva and gastric mucosal samples from 35 client-owned domestic dogs. ey included seven males and 28 females, ranging from seven months to 14 years of age. Twenty-three of these dogs were clinically normal (all females) and in these animals selected, any alteration or underlying disease were discarded through the history, physical examination, complete blood count, serum biochemistry profile (creatinine, blood urea nitrogen, alkaline phosphatase, albumin, and alanine aminotransaminase). In addition, these healthy animals were submitted to sterilization procedure in the counterpart of the upper digestive endoscopy and gastric biopsy sample collection. Twelve had a history of chronic vomiting (seven males and five females; seven dogs with chronic vomiting only, three with chronic vomiting accompanied with weight loss and alteration in appetite, one with chronic vomiting accompanied with diarrhea, and one with chronic vomiting accompanied with hematemesis). ese symptomatic animals were only submitted to upper digestive endoscopy and gastric biopsy sample collection due to gastric symptoms. For selection of the dogs classified with chronic vomiting (related to chronic gastritis) in this study, the inclusion criteria employed was the confirmation of a history of clinical signs such as vomiting episodes accompanied or not with diarrhea, weight loss, and alteration in appetite, for a period superior to 3 weeks [19,20]. Common differential diagnostics of chronic vomiting, such as drugs toxicity, foreign bodies, hepatic failure, or renal disease, were discarded by history, physical examination, complete blood count, serum biochemistry profile (creatinine, blood urea nitrogen, alkaline phosphatase, albumin and alanine aminotransaminase) and ultrasonography and/or radiography previous to procedure and by upper digestive endoscopy on the moment of study. ese animals were arising from the municipalities of Santa Maria and Itaara, located in the central region of the State of the Rio Grande do Sul, Brazil. Preceding sample collection, the dogs did not receive effective eradication protocols for gastric Helicobacter spp. employing the use of the combination of amoxicillin and clarithromycin or metronidazole and proton pump inhibitor (omeprazole or lansoprazole) or famotidine for 14 days [21,22]. However, antibiotics such as ampicillin, amikacin, metronidazole, and sulfamethoxazole-trimethoprim were used per three days in four dogs with clinical signs (dogs number: n.2 has received ampicillin and metronidazole; n.4 has received sulfamethoxazole and trimethoprim; n.21 has received metronidazole; and n.27 has received ampicillin, amikacin, and metronidazole) prior to sample collections.

Sample Collection.
Following a fasting period of 12 hours, all dogs were anesthetically induced with propofol (4 mg/kg) and held in an anaesthetised state with isoflurane and 100% oxygen. Saliva collection was performed with a sterile swab prior to the endoscopic procedure (to prevent cross-contamination of saliva samples with gastric secretions), and the samples were stored in sterile normal saline and frozen at −80°C until further processing. e collection of gastric biopsies was performed with a flexible endoscope (Karl Storz GmbH & Co.KG., Tuttlingen, Germany) and 2.3 mm diameter biopsy forceps (Changzhou Jiuhong Medical Instrument Co. Ltd., Changzhou, China). Following the mucosal evaluation, four biopsy samples were collected from the body and gastric antrum. Impression smears of the two gastric biopsy samples from the body and antrum were prepared on an air-dried slide. en impression smears from the same samples were placed into two tubes (one tube for each gastric zone) containing 10% formalin until further processing. One sample of each gastric zone was submitted for the rapid urease test. For molecular analysis, one sample from each gastric zone was stored in a tube containing sterile normal saline and frozen at −80°C until further processing.
In procedures using additional analgesic drugs and locoregional anesthesia, sterilization in those asymptomatic female dogs was performed postgastroscopy employing the conventional open surgical spaying or laparoscopic-assisted method.

Rapid Urease Test (RUT).
RUT was employed with a Urease test kit (RenyLab, Barbacena, Brazil). Following collection, the biopsy samples for each gastric zone were placed together into a test tube and evaluated for 60 minutes. Colour transformation from yellow to pink within 60 minutes was considered positive, while, no colour transformation within 60 minutes was considered negative.

Cytology.
Following the impression smears of the biopsy samples, the slides corresponding to both gastric zones were stained with a quick panoptic stain. e gastric bacterial colonisation densities (the mean of 10 microscopic fields at ×400) were recorded as follows (−0) no bacteria; (+1 � mild) less than 10 bacteria per field; (+2 � moderate) 10-50 bacteria per field, and (+3 � severe) more than 50 bacteria per field [23,24].

Histopathology.
e formalin-fixed gastric samples for the body and the antrum were sectioned and stained with haematoxylin and eosin and then processed routinely. e gastric samples were evaluated by using the World Small Animal Veterinary Association (WSAVA) gastrointestinal standardisation visual analogue scale [25], where the attributed scores are based on the gastric lesion severity and recorded as follows: (0) lesion not observed; (1) mild lesion; (2) moderate lesion; and (3) severe lesion. e values obtained in the evaluation of the three random fields for each gastric zone were calculated and expressed with the mean of the gastritis severity score.

DNA Extraction and PCR Amplification.
Some animal samples, such as faeces and saliva, had PCR inhibitors. To minimize this issue, we used DNA extraction and PCR protocol used worldwide in the DNA microorganism detection.
us, DNA was extracted from the saliva and gastric biopsies with the DNeasy Blood and Tissue Kit (Qiagen N.V., Hilden, Germany), following the manufacturer's instructions. Total DNA was extracted for each sample and submitted to PCR testing, which used the following oligonucleotide primers (F-5′-AAC GAT GAA GCT TCT AGC TTG CTA-3′; R-5′-GTG CTT ATT CST NAG ATA CCG TCA T-3′), which amplified a fragment of 399 base pairs (bp) for the 16S rRNA gene of the Helicobacter spp. [26]. PCR testing involved 1x buffer of PCR containing MgCl2, 10 mM of dNTPs (0,2 mM of each), 10 pmol of each primer, 1 U of Taq polymerase enzyme (Promega, Madison, Wisconsin, USA), and qsp water; following these conditions, the initial denaturation was 95°C per 5 minutes, followed by 32 cycles at 94°C-30 s; 62°C-30 s to annealing of the initiators and 72°C-30 s for chain extension; and final extension for 3 minutes at 72°C. e amplified product was analysed by electrophoresis in an agarose gel 1% (60V, 1 hour and 30 minutes), using GelRed ® (Biotium, California, USA) and visualised by a transilluminator with ultraviolet light. In all of the amplifications, the positive control was obtained from the gastric mucosal samples from a dog known positive for Helicobacter spp. and the negative controls were ultrapure water samples.

Sequencing and Phylogenetic Analysis.
e amplification products for PCR were submitted in duplicate for nucleotide sequencing via the Sanger method using Prism 3500 Genetic Analyzer equipment (Life Technologies, California, USA). e consensus sequence from the Staden Package program was used to start the nucleotide sequences [27]. Phylogenetic analysis utilised the consensus sequence for each of the amplified samples and the nucleotide sequences of the Helicobacter spp. obtained from the GenBank database (http://www.ncbi.nlm.nih.gov). e sequences were edited and aligned using the BioEdit Alignment Editor software suite, version 7.0.5.3 [28]. e phylogenetic tree was constructed using the MEGA X software [29].

Criteria for Positive Dogs with Helicobacter spp.
A dog was considered positive for Helicobacter spp. when there was a positive result at least one test (cytology; RUT; or 16S rRNA PCR assay).

Statistical Analyses.
Statistical analyses were then performed using the statistical software Action Stat Pro (Estatcamp, São Carlos, Brazil). Kruskal-Wallis test was employed to compare the bacterial colonisation density scores between the regions of the body and antrum. e Spearman test was used to evaluate the correlation between the scores of the histologic gastric lesions and the gastric bacterial colonisation density scores in both of the gastric zones. Fisher's exact test was used to determine if the Helicobacter infection was associated with gastric lesions in the histopathology findings. e significance level used was P < 0.05 for all statistic evaluations.

Discussion
e occurrence of Helicobacter spp. (94.3%) in the dogs belonging to the Brazilian geographic region of this study was higher than that of the dogs from Japan (34.7%) [22], South Korea (78.4%) [30], Denmark (76.7%) [31], Germany (82%) [32], and Portugal (87%) [33]. However, our results were similar with dogs of Poland (96.7%) [11], Iran (95%) [34], Venezuela (95%) [18], and Costa Rica (95%) [35]. Interestingly, these results were different from those observed in human populations, whose occurrence of Helicobacter spp. increased according to the socioeconomic underdevelopment of the region [17], with the aforementioned results demonstrating the high level of geographic variation for these bacteria in domestic dog populations, regardless of socioeconomic level. However, additional data from developed and underdeveloped countries are required in order to correlate the Helicobacter spp. prevalence in populations of domestic dogs from different geographic regions [18,33].
rough the evaluation of gastric bacterial colonisation density scores, 22.8% (8/35) of the dogs showed a mild score, 42.8% (15/35) a moderate score, and 25.7% (9/35) a severe score. us, more than half of the subject dogs were reservoirs of a substantial gastric colonisation density for NHPH.
During the gastroscopy, the main alteration observed was erythema; however, none of the dogs showed the presence of erosions or ulcerations that were described by Kubota-Aizawa et al. [22] and Suárez-Esquivel et al. [35]. Furthermore, some of the gastroscopy findings failed to match the histological findings. For instance, when microscopic lesions were observed, the macroscopic evaluation did not always exhibit alterations; however, these findings are not abnormal [19]. erefore, the diagnosis of gastritis should not be based solely upon gastroscopy visualisation, as biopsy collection is indispensable to confirm the diagnosis of gastritis [36]. e main histologic changes observed in the dogs of this study were the presence of lymphocytic-plasmacytic infiltrates, which were commonly observed when NHPH was present [33]. Nevertheless, NHPH factors regarding the gastric presence and gastric bacterial colonisation density were not found to be correlated with the gastric lesions and severity scores, and this result was similar to other authors' results [9,18]. Moreover, the virulence and pathogenicity could vary between the different Helicobacter species and strains within an NHPH infection [37]. Kubota-Aizawa et al. [22] suggest in vitro cultivation of each NHPH species and experimental infection of canine gastric cell lines or dogs to investigate the pathogenicity of each NHPH species.
is study was the first conducted in Brazil and investigated the presence of Helicobacter in the saliva of dogs. e saliva may exert an important role transmission source of NHPH to humans and other domestic animals [11]. In addition, some authors have suggested that the oral cavity could be a potential niche of Helicobacter spp. colonisation [10]. In this study, the results revealed divergent findings from those of other studies. Commonly, a high prevalence of 71.1% to 100% of Helicobacter spp. was detected in the saliva of canine populations and documented [9,10]; however, the present study observed an occurrence significantly lower, at 25.7% (9/35). Another novel finding was that one dog was positive for NHPH in the saliva and negative in the gastric mucosa, which has not been reported by another study in a canine population.
is observation has been previously   [24]. However, we cannot affirm this based only on a case, but two hypotheses arose with this finding: It is possible that dogs with a gastric absence of Helicobacter spp. can harbour this bacteria in the oral cavity; or, this finding could be resultant of the very low number of Helicobacters distributed in the gastric mucosa of this dog, resulting in the absence of positivity by different methods of detection in the samples evaluated (false negative). e cross-contamination was discharged because of the saliva samples being collected prior to the endoscopic procedure and collection of gastric biopsies. e phylogenetic analysis failed to determine the species of the Helicobacter detected in the individual samples. However, based on the closeness of the samples of this study with strains of Helicobacter species such as H. heilmannii s.s, H. salomonis, H. felis, and H. bizzozeronii, we can infer similarity between our samples with these NHPH species. ese species of Helicobacter are important as they are considered to share close evolutionary relationships and genomic similarities with H. pylori [38][39][40].
is might explain the capacity of these NHPH to readily adapt to the gastric environment of humans and to induce gastric diseases, highlighting the zoonotic nature of these bacterial species [39]. e phylogenetic analysis of saliva samples was only possible to dog number 24, and the other samples were not performed due to a weak fluorescence signal.
Due to Helicobacter infection as an important risk factor, currently in Brazil the high mortality rate associated with gastric cancer, it was the fourth-largest cause of death due to neoplasms for men, and the sixth among women [41] and the prevalence and mortality rates due to peptic ulcers have increased as the population age increases [42]. However, the real percentage of these cases related to NHPH infection is unkown. Notably, as the human population increases their contact with domestic animals, a directly proportional increase in NHPH infected people could result relative to the socioeconomic underdevelopment of the country they inhabit [14]. Furthermore, saliva and faeces of dogs with NHPH may be potential sources of infection for human populations [9,11]. us, this is what motivated the authors to initiate and develop this study, as we were faced with a growing canine population while poor sanitary conditions are still a reality in our country. Due to the zoonotic importance of NHPH bacteria, future cohort studies evaluating the  Veterinary Medicine International prevalence of Helicobacter spp. in the Brazilian human population that could be directly associated with their pets would be extremely relevant in order to elucidate whether NHPH is a concern to public health.

Conclusions
e results of the present study demonstrated a low occurrence of Helicobacter spp. in the saliva, and a high occurrence, predominance, and density of zoonotic importance in the stomach of domestic dogs of the central region of Rio Grande do Sul, Brazil. e saliva of the domestic dogs may be a transmission reservoir of Helicobacter, even in animals without bacterial colonisation.
e Helicobacter gastric presence and gastric colonisation density failed to show an association with the gastric lesions pattern observed during histopathology.

Additional Points
Limitations of is Study. Unfortunately, the limitation in this study was the number of animals employed, which resulted in the limited duration of the study (14 months) and weak routine over this time. Samples from the antrum region were not possible to be collected due to difficult endoscopic access of the pylorus in four animals, and for histopathological analysis, some samples were not viable due to the insufficient amount of tissue or absence of the sample.

Conflicts of Interest
e authors declare no conflicts of interest.

Authors' Contributions
DDGS contributed to design, sample collection, clinical evaluation, and selection of the dogs, tabulation, statistical analysis, data interpretation, and writing of the article; CBEM performed preparation of slides of the impression smear of cytology and interpretation of gastric bacterial colonisation density; JFC performed extraction of DNA, execution of the polymerase chain reaction assay, phylogenetic analysis, and construction of phylogenetic tree; MMF performed histopathological processing and interpretation of gastric lesion severity; LFP performed the endoscopic procedure and the gastric biopsy samples collection; BNA, VM, and OGV performed clinical evaluation and selection of the dogs and sterilization procedure of the female dogs; LRM performed anesthesia of the dogs; STLPF was an advisor and contributed to the design, review, and assistance on the elaboration of the article.