Biodiversity of Potential Vectors of Rickettsiae and Epidemiological Mosaic of Spotted Fever in the State of Paraná, Brazil

Spotted Fever Rickettsioses (SFR) are diseases caused by bacteria of the genus Rickettsia, and are transmitted mainly by ticks. Its eco-epidemiological scenarios vary spatially, and may also vary over time due to environmental changes. It is the main disease transmitted by ticks to humans in Brazil, with the state of Paraná (PR) having the sixth highest number of notified incidences in the country. However, information is lacking regarding the SFR disease cycles at likely infection sites within PR. During case investigations or environmental surveillance in PR for SFR, 28,517 arthropods were collected, including species known or potentially involved in the SFR cycles, such as Amblyomma sculptum, Amblyomma aureolatum, Amblyomma ovale, Amblyomma dubitatum, Amblyomma parkeri, Ctenocephalides felis felis, and Rhipicephalus sanguineus sensu lato. From these Rickettsia asembonensis, Rickettsia bellii, Rickettsia felis, Rickettsia parkeri strain Atlantic Rainforest and Candidatus Rickettsia paranaensis were detected. Ectoparasite abundance was found to be related with specific hosts and collection environments. Rickettsiae circulation was observed for 48 municipalities, encompassing 16 Health Regions (HR). As for socio-demographic and assistance indicators, circulation occurred largely in the most urbanized HR, with a higher per capita Gross Domestic Product, lower Family Health Strategy coverage, and with a higher ratio of beds in the Unified Health System per thousand inhabitants. For environmental variables, circulation occurred predominantly in HR with a climatic classified as “subtropical with hot summers” (Cfa), and with forest type phytogeographic formations. In terms of land use, circulation was commonest in areas with agriculture, pasture and fields and forest cover. Rickettsiae were circulating in almost all hydrographic basins of PR state. The results of this study provide the first descriptive recognition of SFR in PR, as well as outlining its eco-epidemiological dynamics. These proved to be quite heterogeneous, and analyzed scenarios showed characteristics strongly-associated with the outbreaks, with cases presenting clinical variation in space, so illustrating the complexity of scenarios in PR state. Due to the diversity of the circumstances surrounding SFR infections in PR, public health initiatives are necessary to foster a better understanding of the dynamics and factors effecting vulnerability to SFR in this Brazilian state.


Spotted Fever Rickettsioses (SFR) are diseases caused by bacteria of the genus
Rickettsia, and are transmitted mainly by ticks. Its eco-epidemiological scenarios vary spatially, and may also vary over time due to environmental changes. It is the main disease transmitted by ticks to humans in Brazil, with the state of Paraná (PR) having the sixth highest number of notified incidences in the country. However, information is lacking regarding the SFR disease cycles at likely infection sites within PR. During case investigations or environmental surveillance in PR for SFR, 28,517 arthropods were collected, including species known or potentially involved in the SFR cycles, such as Amblyomma sculptum, Amblyomma aureolatum, Amblyomma ovale, Amblyomma dubitatum, Amblyomma parkeri, Ctenocephalides felis felis, and Rhipicephalus sanguineus sensu lato. From these Rickettsia asembonensis, Rickettsia bellii, Rickettsia felis, Rickettsia parkeri strain Atlantic Rainforest and Candidatus Rickettsia paranaensis were detected. Ectoparasite abundance was found to be related with specific hosts and collection environments. Rickettsiae circulation was observed for 48 municipalities, encompassing 16 Health Regions (HR). As for socio-demographic and assistance indicators, circulation occurred largely in the most urbanized HR, with a higher per capita Gross Domestic Product, lower Family Health Strategy coverage, and with a higher ratio of beds in the Unified Health System per thousand inhabitants. For environmental variables, circulation occurred predominantly in HR with a climatic classified as "subtropical with hot summers" (Cfa), and with forest type phytogeographic formations. In terms of land use, circulation was commonest in areas with agriculture, pasture and fields and forest cover. Rickettsiae were circulating in almost all hydrographic basins of PR state. The results of this study provide the first descriptive recognition of SFR in PR, as well as outlining its eco-epidemiological dynamics. These proved to be quite heterogeneous, and analyzed scenarios showed characteristics strongly-associated with the outbreaks, with cases presenting clinical variation in space, so illustrating the complexity of scenarios in PR state. Due to the diversity of the circumstances INTRODUCTION Rickettsioses are zoonosis transmitted by arthropod vectors, and are considered to be emerging or re-emerging in several regions of the world. Their bioagent are Gram-negative, obligately intracellular bacteria of the genus Rickettsia. Rickettsia rickettsii, in the Spotted Fever Group (SFG), is the causative agent of Brazilian Spotted Fever (BSF), which is considered the main human tick-borne disease in the country, due to its high lethality rates. Other SFG rickettsioses, mainly that caused by Rickettsia parkeri strain Atlantic Rainforest and Rickettsia parkeri sensu stricto, are considered secondary due to their milder clinical courses and lack of lethality (1)(2)(3)(4). Within the dynamics of disease transmission, humans are accidental hosts, and the epidemic cycle develops from focal infections, and generally remains highly-localized. However, multiple SFG species can coexist in the same environment, so that several species of ticks that parasitize different mammals can become infected (1)(2)(3)5). Thus, the Spotted Fever Rickettsioses (SFR) enzootic and epidemic cycle in Brazil can be quite complex, since there is a diversity of both infecting agents and a diversity of potential vectors. This results in a variety of potential eco-epidemiological scenarios.
Ticks can act as vectors and reservoirs of SFG rickettsiae. Maintenance of bacterial infection in such arthropods occurs through feeding on a rickettsemic host and transovarian and/or transtadial transmission. Once infected, transmission to vertebrates occurs when the ectoparasite feeds (1,3,6). Thus, rickettsial presence in a given area is related to the existence of species of ixodids susceptible to infection and vertebrates capable of sustaining the tick population. Both of these can vary over time and in space. Additionally, the SFR epidemic cycle can be influenced by the overlap of human activities with tick seasonality (2,7,8).
Currently four SFR epidemic scenarios are known in Brazil: [1] the BSF, whose bioagent is R. rickettsii, with Amblyomma sculptum as its vector in the Southeast and parts of the Southern region; [2] Amblyomma aureolatum vectoring in the metropolitan area of São Paulo (SP), in the Southeast region of Brazil; [3] Amblyomma ovale vectoring R. parkeri strain Atlantic Rainforest in Atlantic Forest fragments in parts of the South, Southeast and Northeast regions of the country; and [4] Amblyomma tigrinum infected with R. parkeri s.s. in the Pampa biome in the South region (2,4,9).
In Brazil, SFR cases are collated by the Sistema de Informação de Agravos de Notificação (SINAN; Notifiable Diseases Information System), which aims to collect, gather and disseminate data among the health surveillance network and other interested parties. In 2001, SFR was considered by the Brazilian Ministry of Health to be a compulsory notification disease and, in 2014, it was included in the list of immediate compulsory notification diseases (within 24 h) (10,11). Currently, there are confirmed cases of SFR in 20 of the 27 federative units within the country, with the majority occurring in the Southeast (1,555/2,127) and South (520/2,127) regions. However, deaths are concentrated in the Southeast region (675/683), and all deaths in the South region (6/683) are recorded in the state of Paraná (PR) (12,13).
In this context, PR was the last state in the Southern region of Brazil to report SFR, with the first human SFR case recorded only in 2006 (14). However, by 2019 there were already 47 confirmed cases across the state, and the data from SINAN (12) in the individual form of epidemiological investigation, from January 2006 to December 2017, show R. rickettsii as a possible bioagent of 8/435 reported cases. Accordingly, PR constitutes a relatively new area in the SFR transmission scenario within the country, having for years been a lacuna within confirmed rickettsia distributions between the states in the Southeast and South regions of Brazil. However, knowledge about the ecoepidemiological aspects of the disease in the state is still poorlydeveloped. Accordingly, the current study aimed to analyze the eco-epidemiological profile of SFR in PR, including identifying and mapping the occurrence of potential vector ectoparasites and infection by rickettsiae, and evaluating the sociodemographic, and environmental factors associated with cases of the disease in the state, as well as levels of relevant government programs.

Study Area
Paraná state, one of the 27 federative units in Brazil (Figure 1A), is located in the northern portion of the Southern region of the country ( Figure 1B). Covering 199,305.236 km², it is composed of 399 municipalities; these being divided between 22 Health Regions (HR) (Figure 1C). Due to its geographical position, the state lies at the intersection between climatic zones, subtropical with hot summer (Cfa), inserted in the Atlantic Forest biome, and temperate with mild summer (Cfb), occurring mostly in the Atlantic Forest biome, with a small portion in the Cerrado biome. According to the 2019 census, PR has a population of 11,433,957 inhabitants, with a demographic density of 52.40 inhabitants/km². It has a Human Development Index (HDI) of 0.749, highest the fifth best in the country (15).

Epidemiological Data
This is a descriptive and exploratory eco-epidemiological study of the notified and confirmed cases of the relationship between exposure and SFR outcome. The analytical units were the municipalities, and the HR in PR, with a time horizon of 11 years (January/2006-December/2017).
In Brazil, every suspected case of SFR requires compulsory notification and investigation, as it is a serious disease (16). Cases were classified as "suspected" whenever the individual presented sudden onset fever, headache, myalgia, and/or had a history of tick bites, and/or contact with domestic and/or wild animals, and/or had been to SFR known transmission areas in the last 15 days, and/or presented maculopapular rash or hemorrhagic manifestations (16). Cases were confirmed when signs, symptoms and epidemiological history matched the suspect case definition, and when infection with SFG rickettsiae was detected by serology (16).
Data were obtained from SINAN, using individual epidemiological investigation forms for notified cases, including notifications of suspected SFR cases (17), made available with protection of patient identities, and complying with regulated national ethical requirements (18). Data were filtered to identify duplications, inconsistencies and typos in the epidemiological records, with subsequent exclusion.
Each confirmed SFR case was georeferenced to municipality, which was then considered as a Probable Infection Locality (PIL). Analyzed outcome variable was the SFR cases notified and confirmed during the studied period. Subsequently, these data were grouped into: mild cases (without hospitalization and death), hospitalizations and deaths. Average annual rates of SFR incidence in HR were then estimated.
As exposure variables, we used environmental characteristics (climatic type, land use and cover and watersheds), as well as sociodemographic and health assistance information [degree of urbanization, Gross Domestic Product (GDP) per capita, coverage of the Family Health Strategy (FHS): Estratégia de Saúde da Família and ratio of beds registered in SUS (Unified Health System: Sistema Único de Saúde) per 1,000 inhabitants]. These variables were described using thematic maps, on which SFR cases were superimposed as points located in the centroids of the municipalities, the diameter of the points being proportional to the number of cases in each municipality. For the sociodemographic and assistance variables, maps were prepared with the distribution in quintiles, and obtained from the SUS Department of Informatics (DATASUS) website (http://tabnet. datasus.gov.br). For environmental variables, the categories available in the data obtained from the Instituto de Terras, Cartografia e Geologia do Paraná (Institute of Land, Cartography and Geology of Paraná) (http://www.itcg.pr.gov.br) website were used.

Potential Vectors of SFR
This information was obtained from the database of the Laboratório de Referência Nacional em Vetores das Riquetsioses (LIRN; Laboratory of the National Reference of Rickettsial Vectors) of the Fundação Oswaldo Cruz (FIOCRUZ; Oswaldo Cruz Foundation), built from samples received and analyzed through the work of the Rede Nacional de Vigilância de Ambiente para Febre Maculosa e Outras Riquetsioses (National Network for Environmental Monitoring for Spotted Fever and other Rickettsial Diseases), from the Ministry of Health (19). This contains records of potential vector ectoparasites from 67 municipalities of PR, collected by the Surveillance teams of the Regional Health Services during investigations of cases and during environment surveillance for SFR, between the years 2013 and 2018. Collections were carried out in the environment (through flannel drag and active search) and on vertebrates (by active search). Collection sites were georeferenced from the HR to which the collection municipalities belonged.
Collected sampling units comprised of specimens from the same host or environment. Ectoparasites were screened and subjected to morphological identification using dichotomous keys (20)(21)(22)(23)(24)(25)(26). Reference specimens were deposited in the FIOCRUZ Coleção de Artrópodes Vetores Ápteros de Importância em Saúde das Comunidades (CAVAISC; Collection of Wingless Arthropod Vectors of Community Health Importance) (Supplementary Table 1), while the remaining specimens (28.276/28.517) were subjected to molecular analysis, resulting in 953 samples. In which, adult specimens were analyzed singletons or in pool of two and, non-adult specimens were pooled according to developmental stage (3-50 larvae and 1-20 nymphs).
Three inter-annual blocks in notifications frequency could be distinguished: 2006 to 2009, 2010 to 2014, and 2015 to 2017 (Figure 3). There were also within-year fluctuations: from January to April there was a constant number of notifications, with a fall in May, followed by an increase in June, a peak in October, and then a fall lasting until December (Figure 4). The time interval between onset of the first symptoms until notification was 7 days for most individuals.
The 2nd HR (Metropolitana) had the highest number of notifications (104/435: 23.90% of total), followed by the 18th HR (Cornélio Procópio) with (42/435: 9.65%) ( Figure 1C) Of the competed notifications (435), 38 were noted as confirmed cases, according to the criteria established by the Ministry of Health (16). Of these, one case was excluded from the analysis because the PIL was located in another state (SP), so that 37 cases were analyzed (23 males, and 14 females: none of whom were pregnant) ( Figure 2B). Confirmed notification rates ranged from 0.23 per 100,000 inhabitants in the 13th HR (Cianorte) to zero in nine other HR (Supplementary Table 2). For confirmation criteria, 81.08% (30/37) of the cases were confirmed by laboratory tests, 13.51% (5/37) by epidemiological clinical means, and in 5.41% (2/37) of the cases this field has not been completed.
The distribution of the 37 confirmed cases, according to first symptoms detection month, fluctuation across the year: a decline in February, with a subsequent increase until April, followed by absence of cases in May, and an increase from June to October there, a peak in November, and from there on, a decrease (Figure 4).
Of the five deaths, three were women and two were men, all living in urban areas. As for PIL characteristics, one (woman) was from a home environment in a peri-urban area and four (two women and two men) from rural areas, two of which (a woman and a man) were related to leisure activities, one (man) to the work routine and one (woman) with field unfilled. In terms of symptoms, all reported fever, myalgia and prostration; while four individuals (two women and two men) had shock and respiratory changes; and two (a woman and a man) had a rash. The median time of disease progression (date of onset of the first symptoms to death) was 13 days (ranging from 6 to 16 days), and the mean time for notification of deaths (date of onset of symptoms and notification) was 15.6 days (ranging from 8 to 28 days).
Mild SFR cases predominated in the most urbanized HR, with higher GDP per capita and FHS coverage ranging from 34.7 to 77.0% (Figures 6A-C; Supplementary Table 2). The most serious cases, which evolved to death, were concentrated in two HR situated in the northeast of the state, 18th HR (Cornélio Procópio: 4 deaths) and 19th HR (Jacarezinho: 1 death), both in the North Pioneiro mesoregion, which are among the less urbanized health regions, with lower GDP per capita and higher ratio of SUS beds per 1,000 inhabitants. The other cases of hospitalization, which did not progress to death, had a sparse distribution within the state (Figure 6; Supplementary Table 2).
Most SFR cases were recorded in Cfa climatic type areas with forest type vegetation (Figures 7A,C). For land use, mild cases and hospitalizations were located in areas with agriculture, pasture-and-fields and forest cover, but all deaths occurred in agricultural areas ( Figure 7B). The Paranapanema I and II basins had most cases of death, while mild cases and hospitalizations occurred in a number of hydrographic basins within the state ( Figure 7D).
A total of 28,517 arthropods were collected, including samples of ticks, fleas, mites and lice. However, some immature specimens were only identified up to genus (Amblyomma and Ixodes) due to the lack of specific morphological identification ( Table 1).
In general, most of the potential vectors were collected in areas of HR with higher urbanization indexes (61.10-95.00%) (Figure 9). For environmental characteristics, the potential vectors were found in both climatic types (Cfa and Cfb) ( Figure 10). However, A. sculptum was widely reported in the northern part of the state (the area with a Cfa type climate), where there are records of R. rickettsii-related SFR deaths ( Figure 10F). The species A. aureolatum, A. longirostre, and A. parkeri dominated in the southeastern region of PR, in higher altitude areas and with Cfb climate (Figures 10A,C,E); A. dubitatum was recorded, predominantly, in the northeast in the Cfa area ( Figure 10B); while the species R. sanguineus s.l., R. microplus, and C. felis felis occurred in both climatic types (Figures 10I-K). As for land use type, potential vectors were collected in areas with agriculture, pasture and field, and forest cover (Figure 11). Potential vectors were recorded in almost all river basins in the state (Figure 12).
Our results point to the occurrence of A. ovale-in the dog-with R. parkeri strain Atlantic Rainforest-on the coast (in preserved Atlantic Forest area); A. parkeri-in humaninfected with Ca. R. paranaensis-in degraded Atlantic Forest; A. dubitatum-with R. bellii-associated with anthropization; A. aureolatum-in human-infected with Rickettsia sp.-in a forest edge area in an urbanized environment; R. sanguineus s.l.in the dog-infected with Rickettsia sp. -in urbanized areas; C. felis felis-in a dog-with Rickettsia SFG-in anthropized areas. Even though there was no infection by R. rickettsii in the analyzed ticks, the occurrence of A. sculptum-its main vector, was recorded in anthropized areas were death had occurred (Figures 8, 13).

DISCUSSION
Ecopedemiological studies are essential for development of an accurate understanding of the different SFR scenarios [e.g., (41)(42)(43)], especially those which consider the complexity of transmission scenarios in the country, or analyze where knowledge is lacking (2,4,9,34,36,(44)(45)(46)(47)(48)(49)(50). This hinders a more contextualized analysis, that could provide the comparative results needed for effective public health action. In PR, the first confirmed SFR case occurred in 2006 (14), resulting in a number of subsequent studies in the state (51)(52)(53)(54)(55)(56)(57)(58)(59)(60). The majority of these have provided basic information concerning rickettsia circulation in a given area, with little direct attempt to study case association or use Surveillance System data. As a result, despite advances in rickettsioses knowledge elsewhere in Brazil, there has been little improvement in the understanding of its enzootic and epidemic cycle in PR. Likewise, epidemiological understanding of SFR in PR remains poorly developed, even though there was an increase in the number of notifications following the first case (Figure 3), signaling greater awareness of the health system for the detection of cases.
The results of the current study allowed us to provide the first description of SFR in PR, which, within a general context, has an eco-epidemiological profile close to that already studied for other endemic areas of Brazil, particularly that of the Southeast region, whose bioagent is R. rickettsii has an eco-epidemiological scenario involving capybara-A. sculptum-sugar cane crops, and in the metropolitan area of São Paulo (SP) dogs-A. aureolatum-Atlantic Rainforest fragments; and also R. parkeri strain Atlantic Rainforest in coastal and inland areas from the same region (dogs-A. ovale-Atlantic Rainforest). However, as the PIL of each confirmed case in PR was considered using a centroid of municipalities, urban areas were used to represent the cases. This can lead to inaccurate spatial representation, and the analysis of related eco-epidemiological variables. For example, using only the urban area as main point of a PIL could underrepresent the relevance of water courses, gallery forests and sugar cane crops to sustain capybaras (amplifier hosts) and ticks (vectors/reservoirs) in the eco-epidemiology of BSF.
Our results show similarities with a recent study of SFR in PR, where data on parasitism of humans by ticks and the distribution of cases and potential vectors were analyzed (43). There are resemblances with: the encounter of A. sculptum and A. aureolatum parasitizing humans (Supplementary Table 3); the encounter of A. sculptum and A. dubitatum in Norte Pioneiro (Figures 9-11); the geographical distribution of mild cases, hospitalizations and deaths (Figures 6, 7).
In this context, the analysis of the frequency of cases in PR shows the presence of three frequency patterns (2006-2009, 2010-2014, and 2015-2017) (Figure 3). These have also been detected in other studies (9,42), and show a direct relationship between public health actions and increases in SFR notifications, especially once Ordinance 1,271 of the Ministry of Health, of June 6, 2014 (11), redefined the classification of the disease as once requiring immediate compulsory notification (within 24 h). Likewise, our results indicate that the temporal distribution of notified cases is greater between August (penultimate month of winter) and December (late spring and early summer), with a peak in October (second month of spring) (Figure 4). This is also the period with the highest number of confirmed cases (Figure 4). Thus, between the final third of winter and the beginning of summer, there is a higher incidence of SFR in PR, which has also been observed for other endemic areas of Brazil (9,61,62). However, we have to consider the diversity of SFR eco-epidemiological scenarios already present in PR, and that these are associated with different species and stages of ticks. Accordingly, in Norte Pioneiro, the region of the state in which all deaths are concentrated, it is likely that the circulation of R. rickettsii occurs associated with A. sculptum nymphs, as these are especially prevalent between the months of June to September (63,64). This coincides with the peak of SFR cases in the country (64). At the same time, there is another scenario, occurring in the PR coastal region, which very likely involves the transmission of R. parkeri strain Atlantic Rainforest by adult stages of A. ovale, which are more frequent in November (65,66). As a result, reducing the SFR morbidity coefficient in PR requires vector control, adopting strategies specific to the species involved in the epidemic cycle, both in terms of the environment and of vertebrate hosts.
Most of the confirmed cases were male, corroborating data from SFR epidemiology studies elsewhere in Brazil (9) and from the United States (CDC, 2015). Our study showed that the highest incidence of confirmed cases was for men and women living in urban areas, but who became infected in rural areas. Therefore, it is important to consider the PIL for a better understanding of the SFR infection scenario. Where, males generally acquired the etiologic agent during leisure activities in rural areas, which seems to indicate during visits to wooded areas, forests, rivers and waterfalls, while female individuals were infected during leisure activities (in the rural area) and conducting domestic tasks in the home environment (in a peri-urban area). This profile has also been observed in elsewhere in Brazil, especially in the Southeast region, the area with the largest number of recorded cases and deaths in the country (67). Likewise, people who self-identified as "white, " constituted the majority of PR SFR patients, a pattern which also seems to be common in other regions of the country (9). Most of the PR population is made up of individuals who declare themselves to be "white." The prevalence of "whites" in the population of PR is probably due to the historical colonization patterns in southern Brazil, which was largely settled by Europeans (Italians, Germans and Poles) during the mid-19th century (68).
As is common in other states in Brazil (9), for most cases analyzed here, serology was used as the confirmatory criterion. However, this diagnostic method does not determine the species of bioagent, and the results must be interpreted within an epidemiological clinical context, since identification of the tick with which the patient was in contact with remains important when confirming SFR infection (42,67,69,70). However, contact with capybaras, present in 5.40% (2/37) of the reports in this study, does not seem to be a relevant factor in SFR infections in PR, though they have been reported as so in RJ (42).
Across Brazil the SFR epidemiological scenario appears to be quite variable. Thus, in the Southeast, severe cases and deaths occur where PILs are mainly in rural or peri-urban areas of the Atlantic Forest and Cerrado biomes, with the epidemic cycle involving the transmission of R. rickettsii by A. sculptum (2,9,45,71,72). In such areas, capybaras and horses are the main vertebrate hosts of this tick; as capybaras behaving as amplifying hosts for R. rickettsii, they are most often involved in the circulation of infections in anthropized environments; while horses, in addition to maintaining tick populations, act as sentinel hosts, because of their prolonged antibody persistence (2,9,44,45,71,73). Infections mainly of men engaged in work or leisure activities in such environments, notably in areas close to water or when working with crops, especially sugar cane (2,9,45,71). Additionally, there has been an increase in the number of cases occurring in peri-urban and urban areas and involving women infected in domestic environments (9,63). This scenario was already well-known from urban and peri-urban areas near Atlantic Forest fragments, where A. aureolatum infected with R. rickettsii reaches households and peri-household environment via dogs that get infested with the tick vector when they go into the woods. However, the involvement of A. aureolatum as R. rickettsii vector is only recognized for a single locus, in the metropolitan region of SP state (2,9,71,74). Thus, it is likely that another tick species, capable of transmitting R. rickettsii, and present in urban and peri-urban environments, may contribute to this increase in cases. In this context, R. sanguineus s.l., a common species in an anthropic environment, involved in the transmission of different species of pathogenic rickettsiae, in various parts of the world (1), is capable of both transstadial and transovarial transmission of R. rickettsii (75). The species has already been detected infected with R. rickettsii in several regions of Brazil (76)(77)(78), and is a potential rickettsia vector for BSF in urban and peri-urban environments. This underscores the need for studies that increase our knowledge of the role of R. sanguineus s.l. in the R. rickettsii epidemic and enzootic cycles in these environments. This scenario cannot be ruled out for PR, where six samples of R. sanguineus s.l. infected with Rickettsia sp., were recorded. Of these, three came from Norte Pioneiro, a HR with confirmed SFR deaths (Supplementary Tables 2, 3).
Further, and pertinent to the SFR cycles, is the fact that, despite the presence of A. aureolatum infected with Rickettsia sp. in the metropolitan region of PR (Supplementary Table 3), this ixodid does not seem to participate in the disease cycle outside SP (2), as it is where there are studies carried out. Therefore, future studies  should pay attention to this data and try to elucidate the possible participation of A. aureolatum in the epidemiological scenario of the disease in PR. Although we did not detect R. rickettsii in our study, which is not uncommon, due to the deleterious effects that this bacterium causes in the tick, even in endemic areas where serious cases and deaths associated with R. rickettsii are known to occur (74,79,80), this scenario resembles that reported for the north of PR, the part of the state where all deaths are concentrated (Figure 2C), and where A. sculptum was found in high frequency ( Figure 10F). Likewise, both data from the present study, and those given by Oliveira et al. (81), point to PR SFR death PILs being commonest in rural areas (with the infection occurring during leisure and work activities), or in peri-urban areas, within a home environment, which agrees with an established BSF scenario (1,2,9). On the other hand, the high levels of R. bellii infection of A. dubitatum, reported mainly from the area where the five deaths in PR were concentrated, may be associated with a rickettsial interference (82,83). The infection of arthropods by rickettsiae pathogenic to humans (e.g., R. rickettsii) can be regulated by infection of non-pathogenic rickettsiae (e.g., R. bellii). This is the result of competition for systemic colonization of ixodids, a process whereby primary infection by one rickettsia species, excludes ovarian infection by another species (84)(85)(86).
The epidemiological scenario of mild and moderate cases in Brazil occurs where the PIL is the rural or urban environment, in a forested area or an edge of the Atlantic Forest biome in the South, Southeast and Northeast regions of the country, especially those areas close to the coast, and the epidemic cycle involving transmission of R. parkeri strain Atlantic Rainforest by A. ovale. The human acquires the bioagent when it comes into contact with the infected tick in the forest, or when its domestic dogs, with free access to the forest, carry the infected A. ovale to a household and peri-household environment, where human parasitism could subsequently occur (1,2,47,87). Such a scenario probably occurs in the coastal region and south of PR, where we found areas of forest close to the urban environment, and where A. ovale infected by R. parkeri strain Atlantic Rainforest was detected [Supplementary Table 3; Figure 13B, (34)]. Given the detection of R. microplus with this rickettsia in the western region of PR (20th HR-Toledo), this tick is potentially implicated in the SFR enzootic cycle and can be used to detect rickettsiae presence and circulation (36). A similar role could be developed for D. nitens, since larvae samples with Rickettsia sp. were found in northwestern PR (14th HR-Paranavaí) (42) (Supplementary Table 3). Although R. microplus and D. nitens show little anthropophilic behavior, and vectorial incrimination reports are currently lacking from the literature, these species can act as indicators of rickettsia circulation in the area. It is important to clarify that the encountered rickettsia DNA does not necessarily confirm the possibility and viability of infection.
Additionally, in the southern part of the state, A. parkeri, a tick of known anthropophilic capacity, has already been found infected by Ca. R. paranaensis, a species genotypically close to R. parkeri, Rickettsia africae and Rickettsia sibirica. All are known to be pathogenic, and produce mild cases, pointing that isolated cases linked to Ca. R. paranaensis could be detected and confirmed in the coming years (35) (Supplementary Table 3;  Figures 8, 13C). Another possible scenario producing mild cases is one in which the PIL is in a rural environment, in a forest area or forest edge of the Pampa biome, in southern Brazil, where the epidemic cycle seems to involve transmission of R. parkeri s.s. by A. tigrinum (4). However, the absence of A. tigrinum in the collected tick samples allows us to rule out this scenario in the currently-studied area.
In Brazil, dogs have been indicated as important indicator animals for a regional SFR epidemiological status, due to their involvement in some rickettsia infection scenarios, directly or indirectly in epidemic and enzootic cycles, or in hosting ectoparasites or rickettsiae (76,(88)(89)(90), as well as the potential risk of transmission to humans, following conveyance of vectors into a domestic environment. Thus, the encounter of C. felis felis, collected from dogs, infected with rickettsia in northern PR (Supplementary Table 3) points to potential involvement of this arthropod and its vertebrate hosts in the rickettsiae enzootic cycle. In this context, the bacteria R. felis and R. asembonensis, genetically close, are reported from fleas and ticks collected from dogs. Even though its pathogenicity is the subject of controversy (91, 92), the bioagent of Flea-borne Spotted Fever, R. felis, has a wide geographic distribution, and is associated with mild cases of rickettsioses, with two confirmed cases in Brazil, based on serological data (93). This fact, associated with the absence of other confirmed cases in the country where this bacterium has been isolated, or has followed all the criteria necessary to define the disease causing bioagent, has resulted in the absence of consensus among researchers concerning these cases in Brazil, and the pathogenicity of R. felis. However, although detected in several species of hematophagous arthropods, including A. sculptum (33,48) and R. microplus in PR (36) (Supplementary Table 3; Figures 8, 13D,E), most reports are associated with Ctenocephalides sp. fleas. Recently described, R. asembonensis, is a bacterium associated with domestic animals, rodents and their ectoparasites. Reports of R. asembonensis from arthropods in the country are recent (34,94,95) (Supplementary Table 3; Figures 8,  13D), and little is currently known about the biology and potential disease capacity associated with this rickettsia in Brazil.
It was not possible to establish the situation in areas in PR where cases occurred with hospitalizations, but without deaths (Figures 6, 7), or even gain a base-line understanding of such scenarios. Although in such areas we detected the presence of ticks (A. aureolatum, A. ovale, A. parkeri, A. sculptum, and R. sanguineus s.l.) already known to transmit SFR bioagents (e.g., R. rickettsii, R. parkeri strain Atlantic Rainforest and Ca. R. paranaensis), we do not know if hospitalizations were due to inherent patient conditions -e.g., comorbidities, genetic characteristics, low immunity -or were specific to the pathogen -e.g., rickettsia species not yet implicated in SFR cases in Brazil, or already implicated, but with different pathogenic profile. Additionally, it is possible to hypothesize that those SFR hospitalizations received correct treatment which thus prevented fatal outcomes. Lack of knowledge of such SFR epidemic scenarios is also not uncommon in other areas of the country (9,34,36,48,49,96), and is probably associated with the lack of adequate and timely investigations of patients and in the associated environment, that would allow the conditioning factors of each case to be characterized and defined.
Collection of ticks Amblyomma incisum and Amblyomma varium; the flea Pulex irritans; the mites Androlaelaps rotundus, Gigantolaelaps sp. and Gigantolaelaps wolffsohni; and the louse Haematopinus suis during surveillance and investigation of SFR cases constitute isolated records and, according to the literature, these arthropods do not participate in the SFR epidemic cycles (8). Likewise, despite the detection of specimens of A. longirostre and Ixodes sp. infected with Rickettsia sp. and Rickettsia SFG, respectively, they appear not to participate in the SFR cycles, but may signal the presence of rickettsial circulation in a region (Supplementary Table 3).
All analyzed scenarios from PR show eco-epidemiological aspects strongly-associated with outbreaks (Norte Pioneiro, metropolitan, coastal and southern regions), with cases varying clinically across the sample area, illustrating the complexity of the SFR regional profile (81,97). SFR remains a challenge for the Brazilian public health system, due to a lack of knowledge among health professionals of its clinical characteristics, and ecoepidemiological aspects associated with infection risk (exposure to potential vector ticks and transmission areas), thus likely resulting in underreporting (1,9). This indicates there is still much to be studied and understood regarding the SFR situation in PR.
Although this study sought a better understanding of the dynamics and vulnerability factors for SFR in PR, and has shown the spatial distribution of endemic areas and highlighted outbreak eco-epidemiological aspects, it is suggested that a number of issues should be addressed to generate an effectively focused capacity to prevent and control such cases: (1) training of health professionals to diagnose suspected cases early; (2) identification of new transmission areas; (3) the need for the National Network for Environmental Monitoring for Spotted Fever and other Rickettsial Diseases to investigate vectors in municipalities with confirmed cases; (4) georeferencing of cases (both suspected and confirmed); (5) improvements in vector collection techniques; (6) full completion of investigation form; (7) permanent health education strategies, to make the general population aware of the inherent risks; (8) additional studies in the vulnerable areas of Norte Pioneiro, metropolitan, coastal and southern regions of the state of PR.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

ETHICS STATEMENT
Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

AUTHOR CONTRIBUTIONS
LD, KB, FR, and GG contributed to the concept, design, and statistical analysis of the work. LD, KB, and GG with the taxonomic identification of ectoparasites. LD and KB contributed to PCR techniques for Rickettsial identification in arthropods. KB with sequencing and molecular analysis of rickettsiae. GG with acquisition of the data. MN in making the maps. KB, FR, GG, EN, and LD with analysis, interpretation, and the drafting the work. All authors contributed revising it critically for important intellectual content, final approval of the version to be published, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

FUNDING
The article is part of the doctoral thesis of LD supported by the Ph.D. scholarship program of the Universidade Federal de Juiz de Fora.

ACKNOWLEDGMENTS
The authors would like to thank the Municipalities and the Secretary of Health of Paraná for the tick collections and for sending samples; the Genomic Platform DNA Sequencing (PDTIS, Oswaldo Cruz Foundation) for support with sample sequencing; Doutora Marinete Amorim for helping with ectoparasite data received by LIRN; Doutor Stefan Vilges de Oliveira for making the data available; and Adrian Barnett for helping with the English of this manuscript.