Molecular Epidemiology of Pathogenic Leptospira spp. Infecting Dogs in Latin America

Simple Summary Leptospirosis is a neglected zoonosis caused by a bacterium of the genus Leptospira. Dogs are highly susceptible to infection, which can range from chronic and asymptomatic to acute and severe disease. Although serosurveys of canine leptospirosis in Latin America are widely published, the gathering of molecular data of strains identified from dogs has not yet been performed. Based on Leptospira spp. DNA sequences collected on digital platforms, we aimed to genetically analyze the strains circulating in Latin America in order to provide an unprecedented overview of the molecular epidemiology of pathogenic leptospires infecting dogs in the region. We found a very high genetic similarity between strains apart serogroup, clinical signs, or geographical localization. The strains were similar to the one that most circulates in humans. In addition to the importance from a zoonotic point of view, this high genetic similarity between strains can, in theory, facilitate the development of an effective broad-spectrum vaccine across the continent, as well as enable the standardization of rapid diagnostic tests. Abstract Canine leptospirosis is a bacterial disease caused by spirochetes of the genus Leptospira. Infections can vary from asymptomatic and chronic infections to clinical acute diseases. The disease is endemic in tropical areas, such as Latin American countries, but a broad understanding of the dynamics of circulation of strains, based on molecular data, has not yet been performed. Based on in silico analyses, the present study aims to analyze the genetic diversity and circulation patterns of haplotypes from pathogenic leptospires infecting dogs in Latin America. DNA sequences were obtained from GenBank platform, curated, and aligned. Genetic distances were calculated, and a maximum likelihood tree and haplotype network were constructed. According to the inclusion criteria adopted, a total of 148 sequences were identified. Most of the records were from Brazil, including sequences from L. interrogans serogroup Icterohaemorrhagiae. Phylogenetic analysis showed a genetically closely related cluster, consisting of a larger haplogroup that includes the reference strain Fiocruz L1-130, known to be the major circulating strain in humans. Moreover, no genetic variations were observed according to clinical history and/or geographical localization. We described the molecular epidemiology of leptospires circulating among dogs in Latin America and demonstrated a very genetically homogeneous group, elucidating its ubiquitous circulation pattern and drawing attention to the important role of dogs in the One Health transmission dynamics of leptospirosis.


Introduction
Leptospirosis is a neglected zoonosis caused by bacteria of the genus Leptospira, and affects domestic and wild animals, as well as humans [1]. Leptospires have a thin and spiral morphology, presenting 6 to 20 µm length and two periplasmatic flagella, producing Infection by L. interrogans serogroup Icterohaemorrhagiae has been assigned to be the major cause of the acute signs, while adapted strains of L. interrogans serogroup Canicola are more frequently associated with mild disease [21,22,29]. Nevertheless, not only the virulence of the strain determines the severity of the infection. Recent evidence demonstrated that the same strain can provoke clinically severe features or induce an asymptomatic infection in different dogs [30]. This study analyzed genetic diversity on the secY gene of L. interrogans serogroup Icterohaemorrhagiae strains obtained from 13 dogs, including animals with signs of acute disease, asymptomatic, and presenting chronic kidney disease. There were no differences between the groups' sequences.
Despite the clinical outcomes, infected animals shed leptospires in the environment, promoting the exposure of other mammals [10,31]. Due to the close proximity to humans, infected dogs play an important role in One Health [32], especially in the cases of asymptomatic harboring [22]. The real impact of canine leptospiral infection and its transmission to humans is poorly understood [32,33]. Nonetheless, reports show that leptospirosis is endemic in tropical areas, such as Latin American countries [21]. Serosurveys of canine leptospirosis in those countries are widely published. In Nicaragua, an epidemiological study from 2007 to 2013 revealed the influence of serogroups Canicola, Pyrogenes, and Louisiana in canine leptospirosis [34]. A serosurvey in Colombia found Autumnalis and Canicola as the most prevalent serovars [35]. Still, seroprevalence studies in Chile reported the presence of Canicola, Pomona, and Autumnalis in both urban and rural areas [36,37]. Isolates were recovered from urine (n = 6), blood (n = 1), and fetus (n = 1) in Argentina. The genetic profile of the strains was identical to L. interrogans serogroup Canicola, L. interrogans serogroup Icterohaemorrhagiae and L. interrogans serogroup Pomona [38]. Recently, a review conducted in Brazil revealed the serogroups Canicola, Icterohaemorrhagiae, and Autumnalis as the most prevalent [29].
As abovementioned, most of studies regarding leptospiral infection in dogs are based on serodiagnosis tests. Serology constitutes an indirect method, and its epidemiological value resides in its ability to predict the circulating serogroups. Even though useful for a collective diagnosis, it presents important limitations for chronic animal leptospirosis, and is not adequate for an individual diagnostic, as serological and molecular results are very often discrepant, and many infected animals may present seronegative. In this context, the complementary use of molecular tools has been encouraged [39]. In the last decades, DNA sequencing has become less costly and more streamlined by automating processes. In addition, data are being increasingly deposited on free-access digital platforms, making it possible to perform advanced studies through bioinformatics tools. Currently, despite a representative sequence bank available, the gathering of molecular data of strains identified from dogs has not yet been performed. As a consequence, Caimi et al. [34] were able to perform a study involving almost 500 leptospiral genomes from worldwide, applying R7L-MLST and obtaining important insights in the phylogenetic and epidemiological scenarios. Therefore, based on in silico analyses, the present study aims to analyze the genetic diversity and circulation patterns of haplotypes from pathogenic leptospires infecting dogs in Latin America.

Collection of Genetic Data
In January 2023, DNA sequences were obtained from the GenBank platform (NCBI/nucleotide) using the following search terms: "Leptospira" AND "dog" OR "Canis familiaris" OR "Canis lupus familiaris" OR "canine" AND "Argentina" OR "Bolivia" OR "Brazil" OR "Chile" OR "Colombia" OR "Costa Rica" OR "Cuba" OR "Ecuador" OR "El Salvador" OR "Guatemala" OR "Haiti" OR "Honduras" OR "Mexico" OR "Nicaragua" OR "Panama" OR "Paraguay" OR "Peru" OR "Dominican Republic" OR "Uruguay" OR "Venezuela". No date interval filter was applied. Sequences referring to any genetic marker were included, as well as complete genomes.
The sequences were carefully manually reviewed and, once confirmed as being pathogenic leptospires obtained from dogs in Latin America, information on species, serogroup, country of origin, and clinical history was collected. Geographical distribution of the genetic data on Latin America was performed on R-studio (v3.6.1.) using the rgdal package.

Phylogenetic Analyses
Sequences from 16S rRNA and secY genes were selected for phylogenetic analysis due to its more representative number of sequences and a well-defined use for leptospiral genetic characterization [39]. Very short sequences were excluded to avoid bias. After the alignment of sequences on ClustalX v 2.0 software [40], a maximum likelihood (ML) tree was constructed using the Tamura-Nei model (TN92) in MEGA X version 10.0.5 software [41], as it was determined to be the best-fitting model of DNA substitution using the Bayesian information criterion. Genetic distances were calculated using the TN92 model on MEGA X.
For better visualization of haplotype distribution, a haplotype network, a widely used approach for analyzing and visualizing the relationships among DNA sequences within a population or species [42], based on secY sequences was constructed through the population genetics software PopART [43], using the media-joining inference method [44]. The reference sequence L. interrogans L1-130 was included in all analyses.

Molecular Epidemiology Based on Phylogenetic Analyses
Sequences from 16S rRNA and secY genes were aligned and submitted to genetic analysis, with a cutoff point for the alignments of 258 bp and 393 bp, respectively. A very low genetic distance was observed in both genetic markers (TN92 < 0.01 ± 0.00), reflecting a higher than 99% similarity between sequences.
The ML phylogenetic tree based on secY sequences (n = 55) confirmed the three circulating species (L. interrogans, L. noguchii, and L. santarosai), and demonstrated a very homogeneous subcluster in L. interrogans, including the reference strain Fiocruz L1-130. No specific clusters were observed according to serogroup, clinical history, or geographical localization, as clearly illustrated in Figure 2.

Discussion
In the present study, we gathered and genetically analyzed the deposited sequences of Leptospira spp. identified in dogs from Latin America. We have searched for sequences of all Latin American countries, most of them belonged to Brazilian studies (81.8%). Nevertheless, this outcome does not reflect that Brazil has a higher occurrence of canine leptospirosis than the others. The high quantity of deposited sequences of Brazilian dogs is most probably a mirror of the research effort in this field. Therefore, the study of canine leptospirosis should be extensively performed in other countries.
The L. interrogans prevalence in dogs was drastically higher than other leptospiral species, just as observed in humans and rats [45,46]. L. noguchii and L. santarosai, although less prevalent than L. interrogans, were also found herein. In Brazil, a strain of L. noguchii serogroup Australis were recovered from a symptomatic male stray dog kidney tissue [47], and two urine samples of asymptomatic dogs were identified as L. noguchii by PCR targeting the gene secY [22]. L. santarosai serogroup Sejroe was recovered from the urine of an asymptomatic dog [48]. In Colombia, L. santarosai was identified by PCR in one unvaccinated dog's urine [49]. Thereby, the presence of pathogenic species of Leptospira other than L. interrogans, although not common, cannot be neglected, since they may act as emerging pathogens to dogs. In addition, L. noguchii and L. santarosai have also been reported in other animal hosts in South America, such as ruminants, opossums, weasels, and armadillos [50][51][52].
In relation to the serogroup distribution, even though only half of the sequences had that information, strains of Icterohaemorrhagiae were by far the major agent of canine infection. In contrast, many serosurveys studies carried out in Latin America frequently reported the presence of strains of Canicola, Autumnalis, and Pomona besides Icterohaemorrhagiae [29,[35][36][37][38], and their real role in canine leptospirosis remains to be elucidated. On the other hand, the number of sequences recognized as the Canicola serogroup was surprisingly low since dogs are known to be the maintenance host of this serogroup [10]. Although few sequences of Canicola were deposited, this serogroup plays a major role in serosurveys conducted in South America [49,53,54]. The fact that only 36.1% of the records were identified to be originated from asymptomatic dogs may be responsible for this low rate of Canicola sequences, this number could be higher if more studies were conducted in asymptomatic or chronically infected dogs. Until recently, infections by incidental serogroups, such as Icterohaemorrhagiae, Pomona, and Australis, were linked to severe clinical illness [11,21]. Our outcomes demonstrate that, despite a great number of sequences being recognized as incidental serogroups, almost one-third of the animals were asymptomatic carriers. Those results reinforce previous studies that demonstrated that, besides the classic host-adapted features between dogs and Canicola strains, dogs can also be infected and shed non-adapted strains and remain asymptomatic [22,30,31].
In fact, the high genetic similarity between the sequences of clinically ill and asymptomatic dogs reinforces the hypothesis that the same strain can lead to different clinical outcomes [30]. However, a better understanding of this question with more accurate and definitive answers will only be obtained with advances in genomics, proteomics, and transcriptomics studies, particularly focused on agent-host interactions.
The genetic characterization of the Leptospira spp. from dogs is discrete. The absence of a standardized genetic marker for sequencing is an important factor in the gap in the molecular characterization of leptospires. In that context, the use of the secY gene as a sequencing target has been encouraged, due to its good taxonomic resolution [39]. Other studies have already performed a wide evaluation of deposited leptospiral secY sequences of other animal species and from different countries [55,56], demonstrating the feasibility of this gene as a target in molecular studies. Although secY sequencing has been encouraged, in the present study, it represented only 35.1% of the obtained sequences. The target 16S rRNA represented more than half of the sequences; this target was one of the firsts used in Leptospira spp. molecular characterization, but its low taxonomic resolution is not useful to differentiate and characterize pathogenic species of the genus [57]. The lipl32 gene, representing 6.8% of the sequences, is specific to pathogenic species of Leptospira and its sequence is very conserved in Leptospira spp., that is why its taxonomic resolution is also low [58]. Nevertheless, even with different gene targets being evaluated, the sequences showed high homology (>99%) between them. It is noteworthy that such similarity suggests the presence of a major well-adapted strain circulating in Latin America.
Regarding the haplotype network, the distribution of the clusters was not influenced by the serogroups, clinical aspects, or the geographic origin of the sequences. The larger haplogroup was composed of more than 40 sequences, all genetically similar to the strain Fiocruz L1-130, (L. interrogans serogroup Icterohaemorrhagiae) known as the major agent of human leptospirosis in Brazil [30,[59][60][61]. The distribution of those clonal strains of L1-130 by Latin America highlights a ubiquitous circulation pattern and a huge concern to public health.
Concerns about the possibility of infectious disease outbreaks and epidemics have grown during the past few decades [62]. Almost all these illnesses are zoonoses and call for a multidisciplinary approach. These worries led to the development of One Health, a collaborative, multidisciplinary, and cross-sectoral strategy to address the threats from Environment-human relations. WHO defines One Health as "an integrated, unifying approach to balance and optimize the health of people, animals and the environment" [63].
Leptospirosis, as one of the most common zoonotic bacterial infections worldwide, encompasses all aspects of a One Health approach. Its understanding involves in-depth knowledge of the transmission processes, animal reservoir hosts involved, environmental sources of the organism, climate conditions that affect the transmission, and the effects of human occupational activities. Maintaining transmission depends heavily on pathogen characteristics, reservoir host animals, and environmental conditions. Dogs may act as sentinels of human exposure to leptospirosis [33] and can harbor and shed pathogenic leptospires. Dogs and their owners may share a similar ecological bioaccumulation habitat, and they may serve as sentinels or reservoirs for a variety of zoonotic diseases. Leptospirosis is one of these, and it poses a risk to both humans' and dogs' lives because it can seriously affect several organ systems and result in death. Particularly in metropolitan areas, the full impact of dogs on the spread of zoonotic illnesses is unknown. In that context, it is known that asymptomatic dogs can persistently shed leptospires in their urine [22,31]. In Latin America, as most countries in the world, licensed vaccines against canine leptospirosis are bivalent, composed of serovars Icterohaemorrhagiae and Canicola [64]. Although vaccination can reduce urinary shedding, even worldwide recommended protocols of immunization do not prevent kidney colonization and vaccinated dogs can act as carriers of leptospires [65]. Similarly, cats infected with Leptospira spp. remain asymptomatic with very low antibody titres, even after vaccination, reflecting a short-term adaptative immune response that tolerates the infection and contributes with the maintenance of the bacteria in rural and urban areas [66].
Currently, both human and canine species have close proximity and share the same housing space, which qualifies dogs as an important pillar in the epidemiology of zoonotic diseases. Latin American countries have an alarming dense population of stray or freeroaming dogs [67] that allied to the tropical environment and high endemicity, can play a crucial role in leptospirosis transmission to humans. Still, little is known about the real impact of the closeness of the canines with humans or the high presence of stray dogs with human leptospirosis features. However, the association of asymptomatic shedding and the infection by strains with high zoonotic potential puts canine leptospirosis as a hazard to One Health.
In this context, the findings of the present study, that strains obtained from dogs in Latin America are essentially the same that determine severe disease in humans and circulates in other domestic and wild species deserves attention. Specifically in the case of dogs and humans, both species share the same environment and epidemiological aspects, reinforce the role of the need for a One Health approach in order to reduce both animal and human cases simultaneously. They are interdependent and, particularly in urban settings where they share the same sanitary conditions and are exposed to the same rodent population, it is not possible to consider the study or the understanding and control of leptospirosis without considering a broad approach to it.

Conclusions
We described the molecular epidemiology of leptospires circulating among dogs in Latin America and demonstrated a very genetically homogeneous group, with a predominance of a larger haplogroup that includes identical sequences to L. interrogans Fiocruz L1-130, elucidating its ubiquitous circulation pattern and alerting to the important role of dogs in the One Health transmission dynamics of leptospirosis.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ani13152422/s1, Table S1: Detailed information about pathogenic Leptospira spp. identified through genetic sequencing in dogs from Latin America. Strains are presented in alphabetical order.

Conflicts of Interest:
The authors declare no conflict of interest.