Entomological survey
Inspection of caught insects revealed, as in previous studies, Phlebotomus mascittii, in very low numbers, but also a single female specimen of Phlebotomus simici NITZULESCU, 1931, namely from Orth an der Donau (48.14462411 latitude, 16.69736534 longitude) in the night of July 8th to July 9th at a local farm (Fig. 1). The CDC light trap baited with dry ice had been put up at the property in a barn with natural floor used for hay storage. Several animals including a dog, cats, chicken, geese, goats, pigs and rabbits were kept at the property. The mean night temperature was 15.6°C and the mean relative humidity (RH) was 62.3% in the respective trap night of July 9th. On July 10th and 11th when no sand flies were in the traps, the mean night temperature was 15.5°C and 13.2°C, respectively and the mean RH was 53.4% and 71.4%, respectively. The village is located in the federal state of Lower Austria in the eastern part of Austria directly along the River Danube, approximately 15 km west of Vienna. The annual mean temperature in Orth an der Donau is 9.9°C and the annual mean precipitation is 627 mm.
Species identification
The specimen was morphologically identified by characters of the pharynx and spermatheca as belonging to the subgenus Adlerius NITZULESCU (Fig. 2). The obtained coxI sequence (GenBank: MN812831.1) was queried against available sequences in GenBank by BLAST and identified as Phlebotomus simici NITZULESCU, 1931 [23]. Sequence identity ranged from 95.99% to 99.85% compared to sequences of Ph. simici originating from Turkey (MN086700.1) and Greece (KU519497.1), respectively. BLAST analysis of the obtained cytb sequence (GenBank: MN812836.1) confirmed species identification and showed 95.0% to 100% sequence identity with sequences of specimens from Crete, Greece (GenBank: MT452061.1) and North Macedonia (GenBank: MT452053.1), respectively. No Leishmania spp. DNA was detected in any of the sand flies by PCR.
Haplotype analysis of Ph. simici based on coxI
Sequences of Ph. simici available in GenBank were edited to compile a dataset of 51 coxI sequences with a length of 551 bp without gaps and stop codons for haplotype analysis (Table 1). 40 haplotypes were identified, defined by 53 variable sites, of which 36 were parsimony informative with an overall haplotype diversity (Hd) of 0.985 and an overall nucleotide diversity (p) of 0.229.
The haplotype of the Austrian Ph. simici specimen (Hap_1) clustered within a conserved European group including haplotypes of specimens from North Macedonia (Hap_2, Hap_3), Thessaloniki, Greece (Hap_3–Hap_7) and Peloponnese, Greece (Hap_8, Hap_9). The haplotype from a specimen originating from Crete, Greece (Hap_10) clustered within the haplotypes of specimens originating from Turkey (Hap_11–Hap_38). A small third group was observed, consisting of both haplotypes of specimens from Israel (Hap_39, Hap_40) (Fig. 3). Analysis of molecular variance revealed 85.6% genetic variation between the three groups and the comparably large genetic distance between the groups was supported by a high FST value (Table 2).
Haplotype analysis of Ph. simici based on cytb
Altogether, seven sequences with a length of 609 bp were included in the analysis (Table 1). Six haplotypes were identified, defined by 34 variable sites of which six were parsimony informative with an overall haplotype diversity (Hd) of 0.952 and an overall nucleotide diversity (p) of 0.322.
The sequence of Ph. simici from Austria was of the same haplotype (Hap_1) as a Ph. simici specimen from North Macedonia, both clustering with the haplotypes of other Ph. simici specimens from North Macedonia (Hap_2) and Peloponnese, Greece (Hap_3–Hap_5). The haplotype from a specimen from Crete, Greece (Hap_6) was clearly separated from all other haplotypes (Fig. 4). As availability of cytb sequences was limited, AMOVA calculation was redundant.
Pairwise sequence comparisons of Adlerius species
Altogether, 82 coxI sequences of Ph. simici and 8 other species of the Adlerius subgenus with a length of 551 bp were included in the analysis (Additional file 1: Table S1). Pairwise distances ranged from 0–18.1%. Hap_1 (Austria) showed the smallest distance (Pd: 0.18%) to Hap_3, which is shared by specimens from North Macedonia and Thessaloniki, Greece, which further corroborated the clustering of the Austrian specimen within the European group in the haplotype network (Additional file 2: Table S2).
When grouping sequences by species, calculated intraspecific mean distances ranged from 0.1–2.4%, the highest being calculated for Ph. simici (Table 3). After a further division into three lineages, namely Europe, Turkey and Israel, mean intraspecific distances were 0.4%, 1.5% and 0.7%, respectively.
Interspecific mean distances between species ranged from 0.8–17.3%. While interspecific distances were low between Ph. simici and Ph. brevis as well as with an unknown Adlerius species from Turkey and Armenia (5.3–6.1%), interspecific distances were high between Ph. simici and other Adlerius species (14.2–17.3%) included. Mean distances between Ph. simici lineages ranged from 2.5–3.9% and 4.5–6.4% between Ph. simici groups, Ph. brevis and Adlerius spp. from Turkey and Armenia (Table 3). The lowest interspecific mean distance of 0.8% was observed between Adlerius specimens from Turkey and Armenia, clearly indicating that these two belong to the same unidentified species.
Nineteen cytb sequences with a length of 609 bp of specimens belonging to the subgenus Adlerius were included in the analysis (Additional file 3: Table S3). Pairwise distances ranged from 0–17.7%. The sequence of Ph. simici from Austria was 100% identical to a Ph. simici specimen from North Macedonia. Pairwise distances to other Ph. simici sequences ranged from 0.5–4.9%, of which the highest was observed to Ph. simici from Crete, Greece (Additional file 4: Table S4).
Intraspecific mean distances were calculated for Ph. simici (1.9%), Ph. halepensis (1.0%) and Ph. chinensis (2.7%), as only one sequence of Ph. brevis was available (Table 4). After splitting Ph. simici into a European lineage and a Turkish lineage including the specimen from Crete, the intraspecific mean distances within the European Ph. simici lineage was 0.6%.
Interspecific mean distances ranged from 9.0% between Ph. simici and Ph. brevis to 15.5% between Ph. halepensis and Ph. chinensis (Table 4). After splitting Ph. simici into a European and a Turkish lineage (including the specimen from Crete), interspecific mean distances were 5.1% between the two groups, 9.0% between Ph. simici European lineage and Ph. brevis and 9.3% between Ph. simici Turkey lineage and Ph. brevis.
Maximum likelihood analysis of coxI
The 82 sequences used for pairwise distance calculations showed 74 unique haplotypes, which were used for ML analysis. Phlebotomus (Transphlebotomus) mascittii GRASSI, 1908 and Phlebotomus (Transphlebotomus) anatolicus KASAP, DEPAQUIT & ALTEN, 2015, as well as Phlebotomus neglectus and Phlebotomus perfiliewi PARROT, 1930, were used as outgroups in two different approaches, respectively. In both approaches, two well-supported major clades were observed, clade 1 comprised Ph. simici, Ph. brevis THEODOR & MESGHALI, 1964, and an unidentified Adlerius species from Turkey and Armenia. Clade 2 comprised all other Adlerius species, namely Ph. chinensis NITZULESCU, 1931, Ph. longiductus PARROT, 1928, Ph. balcanicus THEODOR, 1948, Ph. arabicus THEODOR, 1953, Ph. kyreniae THEODOR, 1958, and Ph. halepensis THEODOR, 1948, (Fig. 5, Fig S1). Calculations resulted in three well-supported lineages of Ph. simici that matched the clustering of the median-joining network. An intraspecific threshold value of 0.7% was used for ABGD analysis, which partitioned the sequences into 11 groups. Calculations were in concordance with ML, with one exception, Ph. simici was split into two hypothetical species, namely Turkey + Israel and Europe. The unknown Adlerius sp. specimens from Turkey and Armenia were shown to belong to one single species and were identified as a sister species of Ph. brevis and together forming the sister group of Ph. simici (Fig. 5).
Maximum likelihood analysis of cytb
The 19 sequences used for pairwise distance calculations showed 15 unique haplotypes, which were used for ML analysis. Phlebotomus mascittii and Phlebotomus anatolicus as well as Phlebotomus neglectus and Phlebotomus perfiliewi were used as outgroups in two different approaches. In both approaches, two well supported major clades were observed, clade 1 comprised Ph. simici, Ph. brevis and Ph. halepensis. Ph. simici and Ph. brevis which further corroborated that they are sister species. Clade 2 comprised Ph. chinensis, which was split into two lineages. An intraspecific threshold value of 1.29% was used for ABGD analysis, which partitioned the sequences into six groups. ABGD grouped all four species as different groups with two exceptions. Ph. chinensis was split into two lineages and Ph. simici from Crete, Greece was computed as a unique Ph. simici lineage, which was in concordance with the ML analysis (Fig. 6, Fig. S2).