Research Article |
Corresponding author: Christoforos Pavlou ( christoforos_pm@hotmail.com ) Academic editor: Bradley Sinclair
© 2022 Christoforos Pavlou, Emmanouil Dokianakis, Nikolaos Tsirigotakis, Vasiliki Christodoulou, Yusuf Özbel, Maria Antoniou, Nikos Poulakakis.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Pavlou C, Dokianakis E, Tsirigotakis N, Christodoulou V, Özbel Y, Antoniou M, Poulakakis N (2022) A molecular phylogeny and phylogeography of Greek Aegean Island sand flies of the genus Phlebotomus (Diptera: Psychodidae). Arthropod Systematics & Phylogeny 80: 137-154. https://doi.org/10.3897/asp.80.e78315
|
The genus Phlebotomus (Diptera: Psychodidae: Phlebotominae) comprises a group of small winged insect species of medical importance. To date, ten species of Phlebotomus are known to be present in Greece; yet their evolutionary history is poorly studied due to the lack of comprehensive phylogenetic and phylogeographic studies. Herein, we aim to clarify the phylogenetic relationships amongst the local species collected from 12 Aegean Islands, Cyprus and Turkey; and to identify which of the palaeogeographic events may have influenced their biogeographic history. Our analyses revealed for the first time the presence of P. cf. major and P. sergenti in the Aegean Islands. All studied local species were retrieved as monophyletic and the mtDNA and nDNA phylogenetic trees indicated a plausible mitochondrial introgression between the closely related species of the P. major complex. From a palaeogeographic viewpoint, the major driving force that shaped the biogeographic history of the studied Phlebotomus species seems to be the dispersal that started in the Oligocene epoch, followed by several speciation events that occurred at the end of Miocene and the Plio-Pleistocene, including multiple dispersal events of Asiatic origin. The Messinian Salinity Crisis, the bimodal Mediterranean climate, and the glacial and interglacial periods were identified as key drivers for the diversification of the local species of Phlebotomus.
Biogeography, Greece, molecular systematics, Phlebotominae, species delimitation
The genus Phlebotomus Rondani & Berte belongs to the subfamily of Phlebotominae (sand flies). They are haematophagous insects and vectors of the protozoan parasites of the genus Leishmania Ross (
Phlebotomus is considered a monophyletic group (
The Aegean archipelago is located in the eastern Mediterranean at the crossroads of three continents (Europe, Asia and Africa). It consists of more than 7,500 islands and islets with various island sizes and geological histories (
The main objectives of our study include the investigation of the phylogenetic relationships amongst the local sand fly species found in 12 Greek Aegean Islands (Fig.
Geographical map indicating all study areas. Α Key map; B Aegean Islands (1: Ikaria, 2: Patmos, 3: Leros, 4: Nisyros, 5: Karpathos, 6: Anafi, 7: Santorini, 8: Folegandros, 9: Milos, 10: Sifnos, 11: Andros); C Crete (12: Botanical Garden-Chania, 13: Fodele-Heraklion, 14: Foinikia-Heraklion, 15: Xerokampos-Lasithi); D Turkey (17: Antalya, 17: Kayseri); E Cyprus (18: Agioi Trimithias-Nicosia).
Sand fly trapping in the Greek Aegean Islands was conducted during previous field work (2016–2019) (
All morphologically identified Phlebotomus spp. are shown in Table
Sand fly species collected from all study areas (
Species/Region | Crete (C: 12–15) | Cyclades (A: 6–11) | Dodecanese (A: 2–5) | North Aegean (A: 1) | Cyprus (E: 18) | Turkey (D: 16-17) |
P. neglectus | √ | √ | √ | √ | √ | |
P. perfiliewi | √ | √ | √ | |||
P. tobbi | √ | √ | √ | √ | √ | √ |
P. halepensis | √ | |||||
P. simici | √ | √ | √ | √ | √ | |
P. creticus | √ | |||||
P. (Adlerius) sp. | √ | |||||
P. papatasi | √ | √ | √ | √ | ||
P. sergenti | √ | |||||
P. similis | √ | √ | √ | √ | ||
P. alexandri | √ | √ | √ | |||
P. killicki | √ |
The sand fly genomic DNA was extracted from the thorax, legs and the anterior part of the abdomen of each chosen specimen using the Qiagen QIAamp DNA micro kit (Qiagen, Hilden, Germany). Double-stranded PCR was performed to amplify partial sequences of two mitochondrial loci (mtDNA) and four nuclear loci (nDNA). These loci were cytochrome c oxidase subunit 1 (COI), cytochrome b (CytB), internal subscribed spacer 2 (ITS2), domain 9 and 10 of 28S ribosomal RNA (28S), elongation factor 1 alpha (EF1-α) and triose-phosphate isomerase (TPI). All PCR reactions had 25μl volume and were performed in T100 thermal cycler (Bio-Rad Laboratories, California, USA). Primers and PCR conditions are described in Table
Locus | Primers | Sequence (5’-3’) | Size (bp) | PCR conditions |
COI | LCO1490 | GGTCAACAAATCATAAAGATATTGG | ~710 | 3mM MgCl, 0.4μM primers, 0.2mM dNTP’s, 1U Taq DNA polymerase |
HCO2198 | TAAACTTCAGGGTGACCAAAAAATCA | (94oC/1min, 42-50oC/1min, 72oC/1min) 35 cycles | ||
C1J1632 | TGATCAAATTTATAAT | ~560 | 3mM MgCl, 0.4μM primers, 0.2mM dNTP’s, 1U Taq DNA polymerase | |
C1N2191 | GGTAAAATTAAAATATAAACTTC | (94oC/1min, 42oC/1min, 72oC/1min) 35 cycles | ||
CytB | CB3-PDR | CAYATTCAACCWGAATGATA | ~550 | 3mM MgCl, 0.6μM primers, 0.2mM dNTP’s, 1U Taq DNA polymerase |
N1N-PDR | GGCAYWTTGCCTCGAWTTCGWTATGA | (94oC/1min, 43-46oC/1min, 72oC/1min) 35 cycles | ||
ITS2 | C1a | CCTGGTTAGTTTCTTTTCCTCCGCT | ~530 | 2.5mM MgCl, 0.6μM primers, 0.2mM dNTP’s, 1.5U Taq DNA polymerase |
JTS3 | CGCAGCTAACTGTGTGAAATC | (94oC/1min, 54-64oC/1min, 72oC/1min) 35 cycles | ||
28S | rc28H | CTACTATCCAGCGAAACC | ~680 | 2mM MgCl, 0.6μM primers, 0.2mM dNTP’s, 1.5U Taq DNA polymerase |
28K | GAAGAGCCGACATCGAAG | Touchdown PCR (94oC/1min, 60+58oC/1min, 72oC/1min) 5+32 cycles | ||
EF1-α | EF-F05 | CCTGGACATCGTGATTTCAT | ~500 | 2.5mM MgCl, 0.6μM primers, 0.2mM dNTP’s, 1.5U Taq DNA polymerase |
EF-R08 | CCACCAATCTTGTAGACATCCTG | (94oC/0.5min, 44-48oC/0.5min, 72oC/0.5min) 35 cycles | ||
TPI | TPI111Fb | GGNAAYTGGAARATGAAYGG | ~460 | 3mM MgCl, 0.6μM primers, 0.2mM dNTP’s, 1.5U Taq DNA polymerase |
TPI275R | GCCCANACNGGYTCRTANGC | Touchdown PCR (94oC/0.5min, 54+50+45oC/0.5min, 72oC/1.5min) 5+5+30 cycles |
Sequence chromatograms were viewed and edited using CodonCode Aligner v.9.0.1 (CodonCode Corporation, Centerville, USA). Multiple sequence alignments for each locus were performed using MAFFT v.7.475 (
The optimal nucleotide substitution model for each locus was identified using PartitionFinder v.2.1.1 (PF) (
All retrieved sequences were submitted to GenBank and their NCBI accession numbers are given in Table S1 of Supporting information. The aligned mtDNA, nDNA and concatenated datasets consisted of 1,080, 1,976 and 3,056 base pairs (bp), respectively. Information on conserved, variable and informative sites is given in Table
Conserved, variable and informative sites at each locus (without the outgroup).
Locus | Total sites | Conserved sites | Variable sites | Parsimony informative |
COI | 619 | 398 | 221 | 211 |
CytB | 461 | 273 | 188 | 174 |
EF1-α | 423 | 294 | 129 | 126 |
TPI | 373 | 281 | 92 | 90 |
28S | 654 | 543 | 111 | 110 |
ITS2 | 526 | 272 | 224 | 204 |
Genetic distances were calculated using the Tamura-Nei model (
We performed a molecular clock test in MEGA before the divergence times estimation analysis was applied to the studied loci. The null hypothesis of equal evolutionary rates throughout the tree was rejected at a 5% significance level (p=<0.05), thus we applied an uncorrelated lognormal relaxed clock for the time estimation (
The coalescent species tree and the estimation of divergence times were calculated using Starbeast2 (
Biogeographic analysis was constructed in Reconstruct Ancestral State in Phylogenies (RASP) v.4.2 (
For conducting species delimitation, we used BP&P v.4.3 (
The genetic distances for mtDNA between the local Phlebotomus spp. ranged from 2.78% to 28.35% and for nDNA ranged from 0.4% to 23.22%. Genetic distances among and within species for mtDNA and nDNA (except ITS2) are given in Table
Genetic distance (%) under the Tamura-Nei model between species. Genetic distance between species for mtDNA (COI & CytB) is given below diagonal and for nDNA (EF1-α, TPI & 28S) above diagonal. Diagonal values represent the genetic distance within species (in parenthesis the nDNA distances). n.a.: not available.
Species | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | |
1 | S. minuta | n.a. (n.a.) | 18.99 | 12.28 | 9.77 | 10.72 | 11.27 | 15.25 | 16.17 | n.a. | 15.15 | 16.88 | 17.22 | 16.03 | n.a. | 20.02 | 16.74 | 13.93 |
2 | P. (Phl.) papatasi | 21.84 | 0.42 (0.29) | 12.89 | 6.81 | 10.06 | 9.57 | 19.70 | 20.32 | n.a. | 17.07 | 18.09 | 23.22 | 19.11 | n.a. | 21.72 | 19.21 | 18.29 |
3 | P. (Art.) alexandri | 20.17 | 15.32 | 0.10 (0.10) | 6.12 | 6.42 | 10.06 | 13.46 | 14.48 | n.a. | 12.72 | 13.55 | 14.45 | 13.50 | n.a. | 14.77 | 15.02 | 14.61 |
4 | P. (Par.) similis | 24.56 | 20.68 | 22.89 | 2.58 (0.01) | 0.40 | 9.07 | 11.50 | 11.29 | n.a. | 11.41 | 10.45 | 10.50 | 10.66 | n.a. | 11.24 | 12.00 | 13.44 |
5 | P. (Par.) sergenti | 24.31 | 19.18 | 20.29 | 13.97 | 1.46 (0.47) | 9.48 | 13.30 | 13.20 | n.a. | 12.91 | 11.87 | 12.02 | 11.63 | n.a. | 13.86 | 13.95 | 14.59 |
6 | P. (Transphl.) killicki | 19.78 | 15.27 | 16.48 | 21.43 | 21.48 | n.a. (n.a.) | 3.11 | 1.42 | n.a. | 2.27 | 6.50 | n.a. | 7.17 | n.a. | n.a. | 8.31 | 7.33 |
7 | P. (Adl.) simici | 19.60 | 17.22 | 15.82 | 22.53 | 21.39 | 13.09 | 1.77 (0.03) | 5.89 | n.a. | 4.80 | 10.88 | 12.22 | 10.62 | n.a. | 12.84 | 10.63 | 9.74 |
8 | P. (Adl.) halepensis | 21.73 | 17.01 | 16.91 | 22.24 | 21.86 | 13.42 | 14.20 | 0.10 (0) | n.a. | 1.59 | 12.07 | 12.18 | 9.86 | n.a. | 11.11 | 10.86 | 9.34 |
9 | P. (Adl.) balcanicus | 20.35 | 17.05 | 17.04 | 21.36 | 23.17 | 11.87 | 12.45 | 8.23 | n.a. (n.a.) | n.a. | n.a. | n.a. | n.a. | n.a. | n.a. | n.a. | n.a. |
10 | P. (Adl.) creticus | 20.34 | 16.00 | 16.48 | 21.87 | 21.27 | 11.73 | 13.15 | 10.46 | 4.44 | 1.20 (0) | 10.61 | 10.19 | 9.03 | n.a. | 10.49 | 9.80 | 8.68 |
11 | P. (Larr.) perfiliewi | 20.59 | 17.17 | 17.18 | 22.32 | 21.50 | 14.42 | 16.27 | 17.46 | 15.17 | 14.34 | 0.95 (0.03) | 3.68 | 3.41 | n.a. | 6.69 | 6.65 | 6.40 |
12 | P. (Larr.) perniciosus | 21.56 | 15.42 | 15.39 | 21.46 | 21.29 | 11.94 | 14.63 | 15.82 | 14.97 | 14.45 | 7.82 | n.a. (n.a.) | 2.66 | n.a. | 8.97 | 8.15 | 9.31 |
13 | P. (Larr.) tobbi | 19.92 | 16.35 | 17.34 | 21.16 | 19.91 | 12.22 | 15.74 | 15.82 | 15.03 | 14.32 | 8.55 | 7.26 | 0.71 (0.09) | n.a. | 8.50 | 8.11 | 7.37 |
14 | P. (Larr.) syriacus | 22.84 | 16.97 | 17.78 | 22.31 | 22.87 | 16.22 | 17.48 | 16.45 | 15.44 | 16.26 | 14.83 | 14.15 | 13.71 | n.a. (n.a.) | n.a. | n.a. | n.a. |
15 | P. (Larr.) major | 27.83 | 21.08 | 18.76 | 26.81 | 28.35 | 15.14 | 17.20 | 17.87 | 16.91 | 16.34 | 15.30 | 13.18 | 13.59 | 3.65 | n.a. (0.78) | 2.04 | 0.94 |
16 | P. (Larr.) neglectus | 20.78 | 17.74 | 16.77 | 22.41 | 21.32 | 16.06 | 16.19 | 16.55 | 15.83 | 15.46 | 14.17 | 13.08 | 13.32 | 5.59 | 2.78 | 1.81 (0.10) | 1.51 |
17 | P. (Larr.) cf. major | 22.43 | 18.29 | 17.92 | 22.21 | 21.04 | 15.98 | 17.15 | 17.94 | 16.16 | 15.32 | 14.29 | 14.11 | 14.16 | 5.86 | 2.98 | 3.58 | 0 (0) |
Both BI and ML analyses on the concatenated dataset resulted into similar and well-supported phylogenetic trees (BI: lnL = –15,458.60; ML: Bootstrap lnL = –15,314.19 and Ultrafast Bootstrap lnL = –15,314.12) (Fig.
The mtDNA and the nDNA datasets produced trees with lnL = –7,489.09 and –7,848.33 for BI, respectively, and; lnL = –7,312.67 and –7,679.85 for Bootstrap ML, respectively and lnL = –7,312.06 and –7,679.21 for Ultrafast Bootstrap ML, respectively. The phylogenetic analyses produced less resolved trees but with similar topologies to that of the concatenated gene tree (Fig. S1 in Supporting information). The most important differences were the positions of the P. major complex, P. sergenti and P. similis. More specifically, according to the nDNA phylogenetic tree, P. sergenti and P. similis form a single well-supported clade. In contrast, the topology of these two species in the mtDNA gene tree was identical to the concatenated phylogenetic tree. Regarding the P. major complex, P. neglectus was resolved as paraphyletic in the mtDNA alignment tree since its lineage contained also the specimens of P. major and P. cf. major (Fig. S1 in Supporting information).
The coalescent species tree analysis resulted in good ESS values (lnL = –16,111.83) and the species tree presented very good posterior probabilities (Fig.
Dated coalescent species tree with posterior probabilities and mean divergence times (above the nodes). The 95% HPD are displayed below the nodes; ancestral geographical distribution is displayed in the up-left corner; letters next to the species correspond to their geographical distribution for the S-DIVA analysis (A: Middle East; B: Aegean Islands; C: East Europe; D: West Europe; E: North Africa).
According to divergence time estimation (Fig.
The results of the biogeographic analysis are shown in Fig.
The results of the species delimitation analyses are given in Table
Species delimitation results assuming 17 species and using different prior schemes. Posterior probabilities are the average of the two runs with different seed number.
Prior scheme | θ~IG(3, 0.2) & τ0~IG(3, 0.2) | θ~IG(3, 0.002) & τ0~IG(3, 0.002) | θ~IG(3, 0.2) & τ0~IG(3, 0.002) | |||
Candidate species | Posterior probability | Posterior probability | Posterior probability | |||
S. minuta | 1 | 1 | 1 | |||
P. similis | 1 | 1 | 1 | |||
P. alexandri | 1 | 1 | 1 | |||
P. tobbi | 1 | 1 | 1 | |||
P. neglectus | 1 | 1 | 1 | |||
P. perfiliewi | 1 | 1 | 1 | |||
P. simici | 1 | 1 | 1 | |||
P. sergenti | 1 | 1 | 1 | |||
P. papatasi | 1 | 1 | 0.99 | |||
P. creticus | 0.98 | 0.98 | 0.99 | |||
P. perniciosus | 0.96 | 1 | 0.92 | |||
P. halepensis | 0.96 | 0.97 | 0.96 | |||
P. killicki | 0.96 | 1 | 0.91 | |||
P. syriacus | 0.95 | 0.99 | 0.90 | |||
P. balcanicus | 0.90 | 0.95 | 0.84 | |||
P. cf. major | 0.89 | 0.86 | 0.92 | |||
P. major | 0.89 | 0.85 | 0.90 | |||
P. major & P. cf. major | 0.11 | 0.14 | 0.08 | |||
P. balcanicus & P. halepensis | 0.03 | 0.03 | 0.02 | |||
P. creticus & P. balcanicus | 0.03 | 0.02 | 0.02 | |||
P. major & P. syriacus | 0.01 | 0.03 | ||||
P. cf. major & P. syriacus | 0.01 | |||||
Number of possible species | Posterior probability | Prior probability | Posterior probability | Prior probability | Posterior probability | Prior probability |
15 | 0.03 | 0.06 | 0.01 | 0.06 | 0.05 | 0.06 |
16 | 0.22 | 0.03 | 0.19 | 0.03 | 0.26 | 0.03 |
17 | 0.76 | 0.01 | 0.81 | 0.01 | 0.68 | 0.01 |
This study is the first multilocus phylogenetic and phylogeographic approach of the Phlebotomus genus in the Greek Aegean Islands. The Aegean area is of great importance regarding biodiversity, biogeography and epidemiology due to its geographic location and history. Our analyses report for the first time the presence of P. sergenti and P. cf. major in the Aegean Islands, which were morphologically identified as P. similis and P. neglectus, respectively by
The interspecific genetic distances of the studied local species of Phlebotomus were similar to those found by
Subgenus Larroussius is closely related to Transphlebotomus and Adlerius, which agrees with the classification of morphological characters by
The first lineage of subgenus Adlerius to branch off was P. simici, while the other three species appear to be closely related. According to our analyses, P. creticus formed a single well-supported clade with P. balcanicus as its closest relative. These results were also observed by
Another important finding of our study is the probable mitochondrial introgression between the species of the P. major complex. In the nDNA gene tree (Figure S1 of Supporting information), three distinct lineages were present, corresponding to the three taxa, and with P. syriacus as the earliest branching lineage. In the mtDNA gene tree (Figure S1 of Supporting information), all taxa were included in a single clade without separating the different species.
According to Bayesian species delimitation analyses, the least supported species are P. balcanicus, P. cf. major and P. major. The taxonomic status of P. balcanicus is unresolved probably because of the only two mitochondrial loci (CytB & COI) that are available in the GenBank database. Further molecular data (nuclear data) of P. balcanicus are needed to resolve its status. The relationship between P. cf. major and P. major is unresolved probably because they may belong to the same species or due to the few molecular data (CytB and EF1-α) available for P. major.
According to our dating analysis, the studied local species of Phlebotomus were separated from S. minuta at 34.37 mya, which is congruent with the suggestion of
Phlebotomus perfiliewi diverged from P. tobbi and P. perniciosus during the pre-evaporitic stage of the Messinian age (7.25–5.96 mya) (
The speciation within the P. major complex appears to have coincided with the glacial and interglacial periods during the early Pleistocene (2.46 to 2.11 mya) (
The subgenus Transphlebotomus is restricted to the Mediterranean basin (
The separation of P. halepensis from P. balcanicus and P. creticus coincided with the MSC (5.96–5.33 mya) (
Finally, the separation of the subgenera Paraphlebotomus, Phlebotomus and Artemievus from each other appears to coincide with the Mid-Miocene Climatic Optimum (~17–15 mya), which represents a geologically warming event (
Under the framework of our regional sampling, all studied taxa were recovered as mutually monophyletic. Phlebotomus sergenti and P. cf. major were recorded for the first time in the Greek Aegean Islands. Furthermore, our results indicated a probable mitochondrial introgression between the species of the P. major complex, while their genetic diversification appears to be low. Phlebotomus creticus was indicated as the sister species of P. balcanicus, with their diversification being the most recent one amongst all studied species. According to our phylogeographic analyses, the palaeoecological events in the Mediterranean region such as the MSC, the establishment of the Mediterranean climate, and glacial and interglacial periods were identified as the major drivers for the diversification of the studied species. Dispersal was the major driving force that shaped the biogeographic history and the current geographical species distributions since most of the species’ diversification was due to dispersal events from Middle East.
Further molecular and morphological research studies are needed for resolving the P. major complex and the relationships between the species comprising it. Likewise, additional molecular data (nuclear DNA) are needed for a more comprehensive study of the relationship and the status between P. balcanicus and P. creticus. Due to the close relationship of these two species, additional studies are required to determine the vector capability of P. creticus. Additional sampling in the Aegean Islands is necessary to locate more specimens of P. creticus and determine its geographical range. This study highlighted the importance of the Aegean Islands and the need for more studies. They may host important sand fly species that play a crucial role in the biodiversity and the epidemiology of leishmaniasis of the area. Finally, further exhaustive sampling in other areas is crucial, in order to collect as many taxa as possible, to clarify the taxonomic status of species complexes and to carry out a more comprehensive phylogenetic study on the genus.
All generated sequences were submitted to GenBank and their NCBI accession numbers are given in Table S1 of Supporting information. All remaining DNA samples and sand fly specimens are deposited in Natural History Museum of Crete (University of Crete).
Specimen collection and morphological identification: CP, NT, VC, YO & MA; molecular procedures: CP & ED; map design: VC; phylogenetic and phylogeographic analyses: CP; data interpretation: CP, MA & NP; project planning: CP, MA & NP; manuscript preparation: CP, MA & NP. All authors have read and approved the manuscript.
We would like to express our gratitude towards the people who helped us with field and laboratory work. Especially, we are thankful to the personnel of the Molecular Systematics and Arthropods laboratories of the Natural History Museum of Crete for their help and assistance with the molecular procedures.
Table S1
Data type: .docx
Explanation note: Information on studied specimens with NCBI accession numbers (n/a: not available).
Table S2
Data type: .docx
Explanation note: The best-fit nucleotide substitution models for each locus/partition selected from PF under the BIC criterion.
Table S3
Data type: .docx
Explanation note: Genetic distance (%) under the Tamura-Nei model between species for COI (below diagonal) and for CytB (above diagonal). Diagonal values represent the genetic distance within species (in parenthesis the CytB distances). n.a.: not available.
Table S4
Data type: .docx
Explanation note: Genetic distance (%) under the Tamura-Nei model between species for EF1-α (below diagonal) and for TPI (above diagonal). Diagonal values represent the genetic distance within species (in parenthesis the TPI distances). n.a.: not available.
Table S5
Data type: .docx
Explanation note: Genetic distance (%) under the Tamura-Nei model between species for 28S (below diagonal) and for ITS2 (above diagonal). Diagonal values represent the genetic distance within species (in parenthesis the ITS2 distances). n.a.: not available.
Figure S1
Data type: .docx
Explanation note: Bayesian inference phylogenetic gene trees (mtDNA and nDNA). Numbers next to the branches represent posterior probabilities (left), bootstrap values (middle) and ultrafast bootstrap values (right).
Figure S2
Data type: .docx
Explanation note: Geographical map indicating the information on dispersal and speciation events between and within areas, as calculated by S-DIVA analysis [A: Middle East (and Cyprus); B: Aegean islands; C: East Europe; D: West Europe; E: North Africa]. *created with mapchart.net.