Genetic diversity of Leptospira strains circulating in humans and dogs in France in 2019-2021

Leptospirosis is a bacterial zoonotic disease. Humans and dogs are susceptible hosts, with similar clinical manifestations ranging from a febrile phase to multiple organ dysfunction. The incidence of leptospirosis in mainland France is relatively high, at about 1 case per 100,000 inhabitants, but our knowledge of the strains circulating in humans and dogs remains limited. We studied the polymorphism of the lfb1 gene sequences in an exhaustive database, to facilitate the identification of Leptospira strains. We identified 46 species-groups (SG) encompassing the eight pathogenic species of Leptospira. We sequenced the lfb1 gene amplification products from 170 biological samples collected from 2019 to 2021: 110 from humans and 60 from dogs. Epidemiological data, including vaccination status in dogs, were also collected. Three Leptospira species displaying considerable diversity were identified: L. interrogans, with eight lfb1 species-groups (including five new lfb1 species-groups) in humans and dogs; L. kirschneri, with two lfb1 species-groups in humans and dogs; and L. borgpetersenii, with one lfb1 species-group in humans only. The lfb1 species-group L. interrogans SG1, corresponding to serovar Icterohaemorrhagiae or Copenhageni, was frequently retrieved from both humans and dogs (n=67/110; 60.9% and n=59/60; 98.3% respectively). A high proportion of the affected dogs developed the disease despite vaccination (n=30/60; 50%). Genotyping with the polymorphic lfb1 gene is both robust and simple. This approach provided the first global picture of the Leptospira strains responsible for acute infections in mainland France, based on biological samples but without the need for culture. Identification of the Leptospira strains circulating and their changes over time will facilitate more precise epidemiological monitoring of susceptible and reservoir species. It should also facilitate the monitoring of environmental contamination, making it possible to implement preventive measures and to reduce the burden of this disease.


Introduction
With an estimated one million cases of severe leptospirosis in humans each year, resulting in 60,000 deaths, leptospirosis is considered one of the commonest zoonoses worldwide.It has been recognized as an emerging global public health problem, because its incidence is increasing in both developing and developed countries (World Health Organization, 2003;Costa et al., 2015;Pijnacker et al., 2016;Bourhy et al., 2017).Leptospirosis is essentially considered a tropical disease, as high temperatures and humid climates favour the survival of the causal bacterium (Evangelista and Coburn, 2010;Costa et al., 2015), and it predominantly affects impoverished populations.Nevertheless, it is also widely reported in temperate areas, including Europe (Goarant, 2016;Pijnacker et al., 2016).Leptospirosis also has a major impact on the health of wild and domestic mammals, and can cause major economic losses in the livestock sector (Noguera et al., 2022).Dogs are particularly susceptible to leptospirosis, and vaccines are available to protect them from disease (Azoćar-Aedo et al., 2014;Ellis, 2015).
The genus Leptospira is currently subdivided into 68 genomic species, including saprophytic and pathogenic species.Infections in humans and animals are caused by only eight pathogenic species: L. interrogans, L. kirschneri, L. noguchi, L. santarosai, L. mayottensis, L. borgpetersenii, L. alexanderi and L. weilli (Vincent et al., 2019).Serological classification based on polymorphism of the lipopolysaccharide (LPS) O-antigen has made it possible to identify more than 300 serovars (sv) grouped into 26 serogroups (sg) (Nieves C et al., 2023).
Infection with Leptospira spp.occurs predominantly through contact between abraded skin or mucous membranes and water or soil contaminated with the urine of infected animals, such as rodents, which are the main reservoir of human leptospirosis (Ellis, 2015).The rat (Rattus spp.) serves as a host for the most widespread serovars worldwide, Icterohaemorrhagiae and Copenhageni (Boey et al., 2019).Other studies have demonstrated the existence of several other hosts, such as mice, voles, and hedgehogs, harboring other Leptospira serovars (Ayral et al., 2016;Izquierdo-Rodrıǵuez et al., 2020;Jeske et al., 2021).Moreover, livestock can also be a source of environmental contamination (Zarantonelli et al., 2018;Sykes et al., 2022).
Human leptospirosis is not a notifiable disease in France.The French National Reference Center (FNRC) conducts passive surveillance by compiling biologically confirmed cases through a network of partners (hospitals and diagnostic laboratories).The incidence of leptospirosis in mainland France is one of the highest in Europe, with approximately 1 case per 100,000 inhabitants or more than 600 cases per year (Bourhy et al., 2017).This disease is associated with working in contact with animals, environmentrelated activities and recreational activities linked to water (Nardone et al., 2004;Guillois et al., 2018;Velardo et al., 2022).
Information about the Leptospira strains circulating in susceptible species is scarce in France and, indeed, in Europe generally.There are several reasons for this.The first one is the laborious nature of Leptospira culture methods, which require fresh biological samples and reference laboratories capable of identifying the strains concerned.Moreover, molecular diagnostic methods, such as Polymerase Chain Reactions (PCR), are gradually supplanting serological tests, which remain the only available epidemiological tool for serogroup identification in Leptospira spp.There is, therefore, a crucial need for a new Leptospira genotyping method suitable for direct use on biological samples that is sensitive, discriminant, simple to implement and inexpensive.Identification of the strains in circulation is essential for diagnosis (to ensure that the diagnostic methods used are capable of detecting these strains), epidemiology (to characterise the reservoirs), surveillance (to detect the occurrence of new genotypes) and prevention (to evaluate the efficacy of vaccines and for the development of new vaccines).
Epidemiological studies of circulating strains are difficult to implement because of the lack of clinical isolates.Pathogenic Leptospira strains are slow-growing bacteria that can be grown only on complex culture media.Current knowledge about the epidemiology of leptospirosis is based on serological results obtained with the reference microscopic agglutination test (MAT), which can be used to identify the infecting serogroup (Faine and Stallman, 1982;Picardeau, 2020).However, this technique has known inconsistencies and weaknesses (Kusum et al., 2005;Smythe et al., 2009).Molecular techniques for studying Leptospira epidemiology have been described, including a core-genome MLST scheme (cgMLST) for isolates (Guglielmini et al., 2019).In the absence of isolates, several alternative methods are available for genotyping, including multispacer sequence typing or MST (Zilber et al., 2014), variable number tandem repeat or VNTR methods (Salaün et al., 2006), and multilocus sequence typing or MLST (Ahmed et al., 2006).Various genes, including the 16S rRNA, lipL32, secY, lfb1, and lic12008 genes can also be sequenced for the direct genotyping of strains present in biological samples without the need for isolation in culture (Merien et al., 2005;Marquez et al., 2017;Santos et al., 2018).
Lfb1 is a putative adhesin of the "fibronectin-binding protein" family (Merien et al., 2000) and lfb1 gene sequences have been shown to be congruent with the genomic species classification of pathogenic Leptospira strains (Merien et al., 2005).Perez et al. showed that the lfb1 locus displayed greater phylogenetic polymorphism than the 16S rRNA gene, making it possible to identify strains down to subspecies level (Perez and Goarant, 2010).This identification method has since been used in numerous epidemiological studies in humans and/or animals (Greǵoire et al., 2020;Soupe-Gilbert et al., 2022).
The FNRC has an exhaustive database of sequences of strains from many regions of the world and from a wide range of hosts.We used this database to evaluate the lfb1 marker for large-scale use for identification purposes.The objectives of our study were to validate the robustness of the lfb1 method for detecting and genotyping Leptospira strains directly from DNA and to propose a new classification system.We then applied this method to the genetic characterization of strains circulating in humans and dogs with acute leptospirosis in mainland France from 2019 to 2021.This work is completed to describe the incidence and geographical distribution in humans, and also the vaccination status in dogs.

Materials and methods
2.1 lfb1-derived phylogeny "reference classification" lfb1 sequences were extracted from the genomes of pathogenic species from the Leptospira cgMLST database (https:// bigsdb.pasteur.fr/leptospira/).This publicly available web-based database currently contains sequences from 834 pathogenic Leptospira strains from different hosts around the world.A clonal group (CG) is defined as a group of cgMLST allelic profiles differing from at least one other member of the group by no more than 40 allelic mismatches over the 545 genetic loci (Guglielmini et al., 2019).
A phylogenetic tree was generated with BioNumerics V7.6 (Applied-Maths, Saint-Martens-Latem, Belgium).The 46 lfb1 sequence strains for the production of the phylogenetic tree are summarized in Table 1.The lfb1 nucleotide sequences have been deposited in Genbank under accession number OR101259 -OR101474.

Samples from humans and dogs
PCR-positive samples collected from humans and dogs in mainland France between January 2019 and December 2021 were included in this study.Cases imported from French overseas territories were excluded from the study.
The samples studied were obtained from i) the routine diagnosis of human samples by real-time PCR (RT-PCR) SYBR-Green targeting lfb1 (Merien et al., 2005), which is performed at the FNRC for Leptospirosis at the Pasteur Institute, and ii) dogs testing positive in routine diagnostic RT-PCR targeting the 16S rRNA (rrs) gene of pathogenic Leptospira strains (Waggoner et al., 2014) at the Laboratory of Leptospira and Veterinary Analysis (LAV) at VetAgro Sup (the Veterinary School of Lyon, France).

Data collection and ethics statement
The human samples and associated data were collected and used in the framework of the surveillance activities of the FNRC for Leptospirosis.These activities are performed in accordance with the mandate awarded to the FNRC by the French Ministry of Health and the French Public Health Code.Dog samples and associated data were collected and sent at by veterinarians from across the country in the context of leptospirosis suspicion.The confirmatory tests were performed by the LAV.
Associated clinical and epidemiological data are entered by clinicians on a document accompanying the samples.These data are often scarce, particularly for human infections.They included the sex and age of the human patient or dog, type of sample, sampling date, geographic information (zip code and region), and vaccination status for dogs.In the absence of geographic data for the patients/ dogs, the address of the laboratory or the veterinary clinic was used assuming that the patients/dogs were exposed in the same region.

Human case definition/serogroup identification
A case was defined as an individual resident in mainland France at the time of infection with clinical findings suggestive of leptospirosis and either a Microscopic Agglutination Test (MAT) titer ≥100 for at least one pathogenic serovar or other laboratory results indicative of leptospirosis (ELISA for IgM, PCR or culture).The panel of strains used for MAT included the serovars Australis, Autumnalis, Bataviae, Canicola, Castellonis, Copenhageni, Cynopteri, Djasiman, Grippotyphosa, Hardjo, Hebdomadis, Icterohaemorrhagiae, Javanica, Louisiana, Mini, Panama, Pomona, Pyrogenes, Sarmi, Sejroe, Shermani and Tarassovi.The infecting serogroup was determined based on the serovar with the highest titer.If the highest titer was recorded for several different serovars, the serogroup was considered to be undetermined.The incidence of annual leptospirosis in humans was calculated with the French population data for 2019, 2020 and 2021 obtained from the National Institute of Statistics and Economic Studies (INSEE: https://www.insee.fr/).
The map of the estimated mean annual incidence of leptospirosis in humans by region and the distribution of the different infecting species-groups were performed with R Studio version 2022.12.0 software, produced with the GeoJSON and Scatter Pie Plot functions.Background map was extracted from https://www.data.gouv.fr/fr/datasets/contours-des-regions-francaises-sur-openstreetmap/and centroid data was extracted from https://www.ign.fr/reperes/centregeographique-des-regions-metropolitaines.

Application of lfb1 genotyping to clinical samples
The lfb1 genotyping method was applied to 170 clinical samples: 110 DNA samples from humans and 60 from dogs testing positive by RT-PCR for lfb1 and 16S rRNA, respectively (Merien et al., 2005;Waggoner et al., 2014).The human DNA samples were obtained from 60 blood samples, 17 urine samples, 4 cerebrospinal fluid samples and 29 DNA extracts from associated laboratories.The dog DNA samples were extracted from 34 blood samples, 21 urine samples and 5 kidney tissue samples.
For both humans and dogs, DNA was extracted with the QIAamp DNA mini kit (Qiagen, Germany), and PCR was performed as described by Merien et al. (2005).The CFX96 realtime PCR detection system (Bio-Rad) was used for qPCR SYBR green assays.The amplification mixture consisted of 0.4 mM primers (lfb1-F 3'-CATTCATGTTTCGAATCATTTCAAA-5' and lfb1-R 3'-GGCCCAAGTTCCTTCTAAAAG-5'), 10 ml of SsoFast EvaGreen supermix (Bio-Rad), and 5 ml sample DNA in a total volume of 20 ml.Samples were amplified with the following program: initial denaturation at 98°C for 2 min, followed by 50 cycles of denaturation for 5 s at 98°C and annealing/elongation for 30 s at 57°C.A 331-bp fragment was amplified corresponding to the complete gene lfb1 for L.interrogans serovar Icterohaemorrhagiae.
All positive PCR products were subjected to Sanger sequencing (Eurofins Scientific, Colonia, Germany and Genoscreen, Lille, France).

lfb1-derived phylogeny "reference classification"
We selected a total of 834 genomes of pathogenic Leptospira isolates (corresponding to 227 different cgMLST clonal groups or CGs), from which lfb1 sequences were extracted for phylogenetic analysis (Table 1).A single-nucleotide polymorphism (SNP) in the alignment between two sequences was considered significant for the creation of Leptospira species-groups (SGs).

Application of lfb1 genotyping to clinical samples
From 2019 to 2021, we collected 110 samples from humans and 60 samples from dogs that had tested positive for leptospirosis by RT-PCR.The lfb1 gene was sequenced from these samples.The sequences obtained and compiled with the classification data proposed above (Figure 1) revealed considerable genetic diversity in samples from humans and dogs, which could be subdivided into three Leptospira species: L. interrogans, L. kirschneri and L. borgpetersenii, corresponding to a total of 11 SGs.We detected L. interrogans in 67/110 (60.9%) of humans and in 59/60 (98.3%) of dog samples, and L. kirschneri in 39/110 (35.5%) of humans and 1/ 60 (1.7%) in dog samples.L. borgpetersenii was detected only in 4/ 110 (3.6%) of human samples.
Based on the classification established above, we detected L. interrogans with eight lfb1 SGs and L. kirschneri with two lfb1 SGs in humans and dogs; L. borgpetersenii was represented by a single lfb1 SG found exclusively in humans.More detailed information is provided in Table 2.
3.3 Epidemiology of leptospirosis in both humans and dogs in France, 2019-2021 3.3.1 Incidence in general human population, seasonality and serotyping 676 human cases in 2019, 450 human cases in 2020, and 708 human cases in 2021 of leptospirosis were diagnosed clinically and biologically in mainland France, corresponding to an average incidence per 100,000 inhabitants of 1.04 for 2019, 0.69 for 2020, and 1.08 for 2021 (Supplementary Table S1).The number of cases clearly increased from July to November in 2019 and 2021, and increased to a lesser extent over this period in 2020 (Picardeau, 2020).Detailed information is available from Supplementary Figure S1.For cases with positive MAT results, the most frequently identified serogroup was Icterohaemorrhagiae (34.4%).More detailed information is provided in Supplementary Figure S2.

Epidemiology of positive samples identified by lfb1 sequencing
Over the study period, the number of human specimens analyzed was highest for 2021 (68 cases, 61.8%), followed by 2020 (23 cases) and 2019 (19 cases).In dogs, the number of cases was highest in 2021 (36 cases, 60%), followed by 2020 (12 cases) and 2019 (12 cases).More detailed information is provided in Table 3.
The month with the largest number of notifications of leptospirosis in humans was August with an average number of 30 cases (27.3%) whereas the numbers of notifications for dogs were highest in July and September, with 11 cases in each of these months (18.3%).For human samples, a male preponderance was observed, with 96/110 (87.3%) of cases obtained from male individuals vs. 14/ 110 (12.7%) from female individuals.More than four fifths (89%) of the patients were adults and the median age was 49 years (5-84 years).A male preponderance was also observed in dogs, with 66.7% of cases (40/60) occurring in male dogs vs. 33.3%(20/60) in females.Just over half the dogs were adults (56.7%) and the median age was four years (0-13 years) (Table 3).
The predominant SG in humans was L. interrogans SG1 -Icterohaemorrhagiae/Copenhageni in the 13 regions of mainland France (Figure 3).The genotypes circulating in human cases were identified in 34 departments (an administrative area equivalent to a county) and 12 regions.The largest number of cases in a region was recorded for Auvergne-Rhone-Alpes in 2021 (Ayral et al., 2016), followed by Pays de la Loire in 2020 (Ellis, 2015), and Bourgogne-Franche-Comtéin 2019 (Goarant, 2016).The genotypes in canine cases were identified in 30 departments and 10 regions, with Auvergne-Rhone-Alpes having the largest number of cases in both 2021 (Ayral et al., 2016) and 2020   (Pijnacker et al., 2016).More detailed information is provided in Supplementary Table S2.
The human population included in this study had not been vaccinated against Leptospira spp.By contrast, 50% (n=30/60) of the infected dogs included in this study had completed the full vaccination protocol for Leptospira.The dogs had mostly been vaccinated with L4 vaccines (63.3% of dogs; vaccine active against serovars Icterohaemorrhagiae, Canicola, Grippotyphosa, and Pomona), L3 vaccines (20%, serovars Icterohaemorrhagiae, Canicola and Grippotyphosa), or L2 vaccines (16.7%, serovars Icterohaemorrhagiae and Canicola).More detailed information is provided in Supplementary Table S3.

lfb1-derived phylogeny "reference classification"
The lfb1 gene is found only in pathogenic species of the P1 clade (Vincent et al., 2019).It encodes a putative adhesin of the "fibronectin-binding protein" family (Merien et al., 2000), suggesting that it may be important for virulence and/or colonization.This gene appears to have changed little during evolution, and is a potentially interesting marker for studies of the genetic diversity of pathogenic strains.Comparative analyses of the lfb1 sequences extracted from available Leptospira genomes can be used to define different clusters or species-groups differing by between 1 and 74 nucleotides over a total length of 281 nucleotides.We found that 119 lfb1 sequences from strains belonging to serovars Icterohaemorrhagiae or Copenhageni isolated from all continents and corresponding to L. interrogans SG1 (cgMLST 6) had identical lfb1 sequences.A single SNP may, therefore, be sufficient to define a new species-group.However, some lfb1 species-groups may contain several serovars or clonal groups, precluding precise identification.For example, L. interrogans SG2 is found in 32 genomes corresponding to different CGs (9, 10, 30, 39, 76, 77, 81 and 400) and different serovars.In this case, identification is less precise and may be a limitation of the proposed method.
The diversity of reservoir hosts for leptospirosis is an important parameter to be taken into account when considering the evolution of Leptospira spp.All these arguments suggest that this gene should be a robust phylogenetic marker.

Application of lfb1 genotyping to clinical samples
SG diversity was greatest in the species L. interrogans (n=13), for which five new SGs were identified in human or dog samples.L. interrogans is isolated more frequently from both humans and animals than the other pathogenic species.The discovery of new SGs was unexpected and suggests that our knowledge of the strains responsible for disease remains incomplete.The Ifb1 genogroups circulating in humans and dogs were identical for L. interrogans SG1/SG5/SG10, with a strong representation of L. interrogans SG1 (sv Icterohaemorrhagiae or Copenhageni), which is responsible for the most severe forms and for which the predominant reservoir in France is the rat (Boey et al., 2019).
The genogroups of species L. kirschneri were frequent in humans and displayed very little diversity (L.kirschneri SG1 and SG4) probably because the L. kirschneri SG1 group contains many different serovars/cgMLST, rendering high-resolution strain discrimination impossible.Similarly, L. interrogans SG5 and L. borgpetersenii SG1 may correspond to several different clonal groups or serovars.The combination of this approach with more refined tools, such as MLST and VNTR, might facilitate more precise identification in such cases.However, our previous analysis of the core genome of strains isolated from several patients (Grillova et al., 2023) showed that strains belonging to cgMLST CG64 (n=10/12) corresponding to L. kirschneri sv Grippotyphosa sg Grippotyphosa (L.kirschneri SG1) and to cgMLST CG72 corresponding to L. borgpetersenii sg Sejroe sv Sejroe predominated in France.The L. kirschneri SG1 genogroup was not detected in any of the dog samples, possibly due to the low level of exposure of dogs to infected environments or a lower susceptibility to L. kirschneri SG1 strains.Indeed, dogs with few symptoms are not presented to veterinarians and recover spontaneously.The L. kirschneri SG4 group was represented by only two strains isolated from a mouse in Bulgaria (identified as serovar Tsaratsovo sg Pomona) and a strain isolated from a French patient in 1990.Samples corresponding to this group were identified only in 2021, and were obtained from five human patients and one dog.The number of samples analyzed was much larger in 2021 than in 2019 and 2020, and this may have made it possible to identify less frequent genogroups.
L. borgpetersenii SG1 was found exclusively in samples from four patients in 2021.It was not found in dogs.It is possible that the Maximum likelihood tree inferred from a Leptospira spp.Ifb 1 partial gene in 46 reference strains.The ID BIGSdb accession numbers are indicated for the reference strains.Further information about the reference strains is provided in Table 1.
human patients were contaminated by cattle, mice or bats, as previously suggested (Greǵoire et al., 2020).
Our data indicate that leptospirosis is common in dogs in mainland France.L. interrogans SG1 (sv Icterohaemorrhagiae/ Copenhageni) and L. interrogans SG5 (sv Bratislava/Lora/Jalna/ Muenchen/Bataviae) predominate in our study and were clearly linked to clinical leptospirosis, whereas L. kirschneri infections may be associated with mild clinical symptoms or possibly be linked to asymptomatic carriage (Renaud et al., 2013;Goy-Thollot et al., 2018).The genotype distribution in dogs were consistent with the findings of previous studies in France (Renaud et al., 2013;Ayral et al., 2014;Andre-Fontaine and Triger, 2018;Hidalgo Friaz et al., 2023).Prevention is the best way to protect dogs from leptospirosis.It can be achieved through vaccination, avoiding contact with contaminated water and infected animals.In this study, the human patients were not vaccinated, because vaccination against leptospirosis is reserved for individuals in certain high-risk professions in France.By contrast, the leptospirosis vaccine is one of the recommended the core vaccines for dogs in France.Nevertheless, 50% of dogs included in this study were diagnosed with leptospirosis despite complete vaccination annually.Several studies have reported similar results, demonstrating the relative nature of the efficacy of current vaccines in France (Ayral et al., 2014;Andre-Fontaine and Triger, 2018;Bertasio et al., 2020).However, quadrivalent vaccines (L4, directed against serogroups Canicola, Icterohaemorrhagiae, Grippotyphosa and Australis) would be a more appropriate vaccine for preventing severe forms and deaths from leptospirosis (Klaasen et al., 2022;Hidalgo Friaz et al., 2023), provided that vaccination protocols are properly respected by owners and veterinarians.

Incidence of human of leptospirosis
The incidence of leptospirosis in humans over the three years of the study was about 1 per 100,000 inhabitants, but was lower, at only 0.69 per 100,000 inhabitants, in 2020.This decrease is probably related to the COVID-19 pandemic, during which recreational activities were limited and the number of imported cases was decreased by travel restrictions.Similarly, little seasonality was observed in 2020.Incidence was highest in regions with livestock production, high rainfall, and large rivers.However, some fluctuation was observed from year to year and the data may not necessarily reflect the situation on the ground at a particular time point.For example, people may become infected during their summer vacation but are not diagnosed until they return to their home region.
The French Ministry of Health plans to declare leptospirosis to be a notifiable disease in the near future.This initiative will improve the follow-up of cases and provide associated epidemiological data (number of cases, disease severity, site of exposure, mode of contamination).The FNRC will play an important role in confirming the biological diagnosis and tracking the serogroups/ genogroups involved in these infections.The follow-up of serogroups based on MAT shows that Icterohaemorrhagiae is the most frequently encountered serogroup, consistent with the results of Ifb1 genotyping.However, very little information is available for  the other serogroups.The regional distribution of origin of samples and of the lfb1 genogroups to identify in mainland France appears to be generally homogeneous for human infection, but not for dogs, for which samples were mostly obtained in the Rhone-Alpes region, in which the LAV laboratory location.More regular surveillance of canine cases in different French regions over a period of several years would be useful as it would provide a more precise idea of the Leptospira genotypes circulating in this species, and would make it possible to establish the epidemiological basis of transmission between humans and dogs.Such surveillance would also make it possible to update preventive measures in humans and animals.

Conclusion
We provide here the first global description of the Leptospira strains responsible for acute infections in humans and dogs in mainland France.We demonstrate that polymorphism of the lfb1 gene is a robust method to provide rapid identification using biological samples.This tool has enabled us to also identify five lfb1 L. interrogans species-groups never before described.The availability of more precise epidemiological data in the future should facilitate the identification of sources of animal and environmental contamination, making it possible to establish public health control interventions.

FIGURE 2
FIGURE 2 Phylogenetic tree based on fbl 1 sequences from human and dog samples.Maximum likelihood tree inferred from Leptospira spp.lfb 1 partial gene polymorphism in clinical specimens and reference strains.Gray boxes indicate reference strains, orange boxes indicate infected humans and blue boxes indicate infected dogs.The colors on the right (green, red, and purple) correspond to three different species of Leptospira.The ID BIGSdb numbers of the reference strains are indicated.Further information about the reference strains is provided in Table2.

TABLE 1
List of Leptospira species groups (SGs) defined on the basis of lfb1 sequence polymorphism.

TABLE 2
lfb1 species-groups identification for positive human and dog samples, by year.