Understanding the factors that determine the emergence of anthroponotic cutaneous leishmaniasis due to Leishmania tropica in Morocco: Density and mitochondrial lineage of Phlebotomus sergenti in endemic and free areas of leishmaniasis

Anthroponotic cutaneous leishmaniasis (ACL) due to Leishmania tropica is spreading to new areas in Morocco. Exposure to the vector, Phlebotomus sergenti , is the only proven risk factor. Our objective was to compare the densities and genetic characteristics of P. sergenti populations in two nearby localities in Morocco, one in an ACL endemic area (El Borouj) and another in a nonendemic area (Sidi Hajjaj). P. sergenti density was significantly higher in the endemic area than in the nonendemic town ( p = 0.032). A different predominant P. sergenti mitochondrial lineage was evidenced in each one of the two localities, and for the first time, the P. sergenti lineage acting as a vector of L. tropica has been identified. Bioclimatic differences were detected between both localities. In conclusion we found differences in both the density and the mitochondrial lineage of P. sergenti populations that may explain the different epidemiological situation. Given that the density of P. sergenti in the locality without ACL cases seems sufficient to allow transmission, the main factor that would justify its nonendemic character could be the absence of P. sergenti Lineage IV, which seems to prefer warmer and drier climates. infected belong to the preva-lent P.sergenti lineageinElBorouj.Thisisthefirsttimethatamitochon-drial lineage of P. sergenti is involved as a vector for L. tropica . A local increaseintheabundanceofthis P.sergenti lineagethatseemstotrans-mit L. tropica more efficiently, could explain the emergence of ACL in El Borouj and its absence in Sidi Hajjaj.

, and Phlebotomus similis is considered a probable vector on the island of Crete (Ntais et al., 2014). Although P. sergenti is believed to have a marked preference for semi-arid habitats (Boussaa et al., 2009), this species exhibits a wide ecological plasticity in Morocco making increased vector surveillance essential to prevent and control leishmaniasis outbreaks. In emerging ACL Moroccan foci, P. sergenti density varies from 4 to 16 specimens/m 2 (Ramaoui et al., 2008) and relative abundance ranges from 12.8% to 76.7% (Boussaa et al., 2009).
The World Health Organization included Morocco among one of the 12 high-burden countries for CL (WHO, 2016). There are three endemic Leishmania species in Morocco: Leishmania major, L. tropica, (both dermotropic) and Leishmania infantum (mainly viscerotropic).
Leishmania tropica has the widest geographic distribution (Ministry of Health, Morocco, 2016;Mouttaki et al., 2014) but until 1989, ACL had been mainly reported in hypoendemic rural foci scattered around the sub-arid area of central Morocco. Later, ACL emerged in several northern, central and southern provinces of the country, initially as new outbreaks and then establishing endemic foci that highlighted the expansion of this Leishmania species (Ajaoud et al., 2013). The first CL case in Settat province (central Morocco) was detected in El Borouj locality in 2006, preceding an epidemic outbreak and then becoming endemic (Amarir et al., 2015;Gijón-Robles et al., 2018). Currently, El Borouj is the only active CL focus in the province of Settat (Ministry of Health, Morocco, 2016).
The identification of factors that determine the emergence and expansion of ACL is required to develop better interventions for this largely neglected disease. We showed that differences in the exposure to the L. tropica vector, reflected by differences in P. sergenti density in the households, was the only factor associated with CL cases in the El Borouj focus (Gijón-Robles et al., 2018).
On the other hand, three of the four mitochondrial (mt) lineages previously reported within P. sergenti are present in Morocco (Barón et al., 2008;Merino-Espinosa et al., 2016;Yahia et al., 2004). Phenotypic differences of biomedical importance may exist between these mitochondrial lineages, thus population genetics could help to assess the threat of the geographical expansion of ACL. Therefore, our objective was to compare the densities and genetic characteristics of P. sergenti populations in two nearby localities in Morocco, one within an ACL endemic area (El Borouj) and another in a nonendemic region (Sidi Hajjaj).

Sand fly collection and species identification
Sand flies were caught in both localities using CDC light traps inside households and sticky papers outside dwellings, from June 20 to July 10 and from September 20 to October 10, 2015. In El Borouj, houses with and without ACL cases were sampled. One to two CDC traps were set in each selected house for one night under favourable weather conditions. Sticky traps consisted of 21 × 29.5 cm sheets of papers covered in castor oil, 9-17 traps were set the same day in adult sand fly resting places (holes in house walls and other nearby walls) and left for 4 days. The captured sand flies were stored in 70% alcohol. Males and females were separated and morphologically identified using taxonomic keys (Benabdennbi et al., 1999;Berchi et al., 2007;El Sawaf et al., 1989;Gil Collado et al., 1989;Leger et al., 1983;Rioux & Golvan, 1969;Rioux et al., 1978;Sáez et al., 2018). The specimens were placed in Marc André solution and heated to boiling point, and finally mounted on slides under a coverslip using Berlese solution. The genitalia of P.
sergenti specimens were individually removed and mounted on slides under a coverslip for morphological identification whereas the rest of the body was stored at −20 • C for DNA extraction.
The gonotrophic cycle of the female sand flies was categorised as blood-fed, non-fed or gravid. Density (sand flies/trap), relative abundance (% specimens of a given species/total sand flies) and frequency (% positive sampling stations for a given species) data were estimated by species.

Mitochondrial lineage determination by mt DNA Cyt b PCR-RFLP
Polymerase chain reaction (PCR) was used to amplify a 550-bp fragment containing the 3′ end of the Cytochrome b mitochondrial gene (mt DNA Cyt b) following the methodology described by Esseghir et al. (1997).
Restriction fragment length polymorphism (RFLP) was carried out through the digestion of a 550-bp mtDNA Cyt b fragment with Hae III (Thermo Scientific, Germany). The reaction was performed at 37 • C for 10 minutes in a 20 µL total volume, containing 16 µL of PCR product, 2 µL of enzyme (10 U/µL) and 2 µL of standard buffer (10×). The

2.5
Detection of Leishmania tropica DNA

Bioclimatic differences between El Borouj and Sidi Hajjaj
In order to investigate the possible association between bioclimatic characteristics and the presence/absence of ACL in the two studied localities, a logistic regression analysis was carried out including the locality as a dependent variable and each bioclimatic variable under study as an independent variable. The bioclimatic data analysed were as follows: mean monthly average temperature (from January to December), maximum monthly average temperature (from January to December), minimum monthly average temperature (from January to December), maximum annual average temperature, minimum annual average temperature, monthly precipitation (from January to December), annual mean temperature (BIO1), mean diurnal range (mean of monthly -max temp -min temp-)(BIO2), isothermality (BIO2/BIO7) Average values (95% confidence interval) and minimum and maximum values of each of the bioclimatic variables from El Borouj and Sidi Hajjaj are shown as supplementary data.   (Table 1). Regarding differences between capture periods the intradomiciliary density of P. sergenti was significantly higher in June in El Borouj (p = 0.012) and in Sidi Hajjaj (p = 0.020).

Sand fly fauna and abundance by species
Blood-fed, non-fed and gravid P. sergenti females were found both in intradomiciliary and peridomiciliary captures in both localities; density values were higher in El Borouj (Table 2).

Leishmania infection rate in the vector
L. tropica DNA was not detected in any of the 84 female P. sergenti captured in Sidi Hajjaj.

Mitochondrial lineage determination by mtDNA Cyt b PCR-RFLP
The mitochondrial lineage was identified in 81 male and female P. sergenti, 41 from El Borouj and 40 from Sidi Hajjaj. The results are shown in Table 3.

Bioclimatic differences between El Borouj and Sidi Hajjaj
Bioclimatic data were collected from 199 georeferenced points, 185 houses with ACL cases and 14 houses in Sidi Hajjaj. Table 4

DISCUSSION
In Morocco, ACL due to L. tropica is transmitted by P. sergenti which has a large geographic distribution probably related to the wide ecological plasticity of this vector (Boussaa et al., 2009;Ramaoui et al., 2008;Rioux et al., 1986). CL due to L. tropica is an emerging disease even though the geographical extension of the vector is greater than that of the parasitic protozoan, and the identification of factors for parasite expansion is essential for effective disease control. Phlebotomus sergenti density and genetic characteristics were investigated as determining factors for the existence of ACL transmission. Comparative intradomiciliary and peridomiciliary sand fly captures in the ACL endemic locality of El Borouj and the non-endemic locality of Sidi Hajjaj were made using CDC light traps and sticky papers.
Phlebotomus sergenti density in Sidi Hajjaj was lower than that of the endemic locality, both in peridomiciliary settings and within households. Interestingly, P. sergenti was the most abundant and densest species within households in Sidi Hajjaj, however it was the fourth species outdoors, after S. minuta and the L. infantum vectors, P. perniciosus and P. longicuspis (Table 1).The relative abundance of P. sergenti males and females varied between the trapping methods since the males were more abundant in the sticky papers. In both localities, the density of P. sergenti females was higher in the intradomiciliary captures carried out in June and non-fed, fed and gravid females, were found (Table 2).
Although to date no ACL cases have been diagnosed in Sidi Hajjaj, these sand fly density figures seem sufficient for the maintenance of L. tropica transmission (Barón et al., 2013;Ramaoui et al., 2008;Rioux et al., 1986) and would make this locality susceptible to the establishment of an ACL transmission cycle. Over the last few decades, L. tropica foci have spread to several regions of Morocco including those where CL caused by L. major or L. infantum has been reported, which shows the changing geographical patterns of this species (Baghad et al., 2020).
The growing mobility of humans raises the possibility of new emerging foci in areas where P. sergenti populations are well established. Kholoud et al. (2020) (Barón et al., 2008(Barón et al., , 2013Merino-Espinosa et al., 2016). The existence of differential ecological traits between P. sergenti mitochondrial lineages has been pointed out: Merino-Espinosa et al. (2016) found that Lineage I appear to have adaptive advantages represented by a wider tolerance to temperature and altitude changes that would make it better suited to lead a geographical expansion into Europe. Similarly, there are bioclimatic differences between El Borouj and Sidi Hajjaj ( Table 4) that could explain the over representation of lineage IV in El Borouj, which is warmer and drier, and its absence in Sidi Hajjaj.
In general, rainfall is greater in Sidi Hajjaj than in El Borouj with a strong seasonality from July to October. In contrast, temperatures are almost two degrees higher in El Borouj throughout the year; however the differences were only significant between the maximum tempera-tures of July. The coefficient of isothermality reflects well these differences in temperature between both localities, being two points higher in El Borouj. Environmental humidity and temperature are two determining factors in the distribution of P. sergenti (Barón et al., 2013) and within this species there appear to be lineages with different ranges of tolerance. Particularly in El Borouj, the presence of water wells was independently associated with a greater P. sergenti density in the households (Gijón-Robles et al., 2018).

CONCLUSION
The density and genetic background of the L. tropica vector, P. sergenti, seem to play a significant role in the prevalence of ACL in Morocco. Differences in both the vector density and its main mitochondrial lineage in two localities, one endemic and the other free of the disease, may explain the different epidemiological situation. Given that the P. sergenti density in Sidi Hajjaj seems sufficient to allow transmission, the main factor that would explain the absence of ACL cases could be the lack of P. sergenti lineage IV, which seems to prefer warmer and drier climates.

ACKNOWLEDGEMENT
This study was funded by the University of Granada (Centro de Iniciativas de Cooperación al Desarrollo, CICODE, 2013). Funding for open access charge: Universidad de Granada/CBVA.

CONFLICT OF INTEREST
The authors declare no competing interests.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.