Background

Anopheles maculatus (sensu lato), in the Neocellia Series [1] of the subgenus Cellia, is a widespread species in Asia, ranging from the Indian subcontinent to Southeast Asia and southern China [2,3,4,5,6]. In Indonesia, this species is widely distributed in the western part of the archipelago extending to Weber’s Line, a hypothetical biogeographical separation between Sulawesi and the Maluku Islands chain [7]. Anopheles maculatus has been recorded in Sumatra, Java, Kalimantan, Bali, Lesser Sunda Islands including East Timor (Democratic Republic of Timor-Leste), and Sulawesi [8].

Prior to the cytogenetic identification of different chromosomal forms, An. maculatus was regarded as a single taxon [9, 10]. Currently, based on phenotypic characteristics, crossmating experiments, cytogenetic and molecular analyses, the Maculatus Group [11] is divided into two subgroups and nine species [3, 10, 12]. The subgroups are differentiated by distinct morphological characters. The Maculatus Subgroup [13] includes An. maculatus (sensu stricto) Theobald, 1901 and Anopheles dravidicus Christophers, 1924, while the Sawadwongporni Subgroup [13] comprises Anopheles sawadwongporni Rattanarithikul & Green, 1986 [9], Anopheles notanandai Rattanarithikul & Green, 1986 [11] and Anopheles rampae Harbach & Somboon, 2011 [14]. The four other species in the group include Anopheles greeni Rattanarithikul & Harbach, 1991, Anopheles dispar Rattanarithikul & Harbach, 1991, Anopheles willmori James, 1903 and Anopheles pseudowillmori Theobald, 1910 [12, 15, 16].

The Southeast Asian mainland presents the highest diversity of the Maculatus Group, with seven species present in Thailand [3, 17, 18]. Anopheles greeni and An. dispar appear restricted (endemic) to the Philippines [15]. Five species are found in China excluding An. notanandai and An. rampae [3, 19]. In Vietnam, four species [An. maculatus (s.s.), An. pseudowillmori, An. sawadwongporni and An. rampae] are present [20,21,22]. Until recently, only An. maculatus (s.l.) [presumed (s.s.)] was reported in Indonesia [8]. A second species has been suspected present in the Kulon Progo District area in Central Java since the late 1990s (MJB, personal communication). This putative, as yet undescribed species was recently reported from material derived from a continuously colonized strain reared [23] over three decades at the Indonesian Ministry of Health Institute for Vector and Reservoir Control Research and Development, a component of the National Institute of Health Research and Development (NIHRD-IVRCRD) and described in this work [24]. Anopheles maculatus has long been considered a major malaria vector in West (peninsular) Malaysia [25], and areas of Sumatra and Java, Indonesia [26,27,28,29,30], predominately in rural, forested areas [31]. Numerous instances of natural malaria plasmodia infections in An. maculatus have been reported in Indonesia [31]. Infection indices have varied from 2.83% in Kisaran (Sumatra) to 3% in Central Java, 17% in Londut (Sumatra) and 11% in Riau Province (Sumatra) [26, 32]. This species is a major public health concern in the Menoreh Hills region, which includes the Kulon Progo District near the border of Central Java Province and the Special Region of Yogyakarta. It is also reported as a major malaria vector in southern Sumatra (Tenang) [33,34,35,36]. Interestingly, although present in Kalimantan, Sulawesi, Bali and the larger islands in the Nusa Tenggara (Lesser Sunda Islands) region, An. maculatus has either not been reported as a malaria vector or is an epidemiologically insignificant species in these areas [37].

We analyzed the diversity and phylogeny of An. maculatus samples collected in different locations and islands in Indonesia. We also analyzed the relationship of the proposed novel species present in Kulon Progo District and reared at NIHRD-IVRCRD with other members of the Maculatus Group to derive its putative origin.

Methods

Mosquito collections and identification

Adult mosquitoes were collected from field settings using standard procedures for human-landing and cattle-landing methods [38] in six provinces of Indonesia between 2012 and 2018. Sampling locations included Cilacap, southern Central Java (samples C1 and C2; October 2011), Belu, West Timor, East Nusa Tenggara (samples NT64 and NT 101; November 2011), Ogan Komering Ulu, South Sumatra (samples S9 and S33; October 2011), Sebatik Island, northern Kalimantan (samples N2 and N44; November 2011), Purbalingga, Central Java (sample P1; September 2011), Kulon Progo, Central Java (samples KP10 and KP72; November 2013) and the NIHRD-IVRCRD laboratory, Salatiga (samples 1x, 2M and 4M; October 2018) (Fig. 1; Table 1). Anopheles maculatus samples were initially identified using morphological criteria [39]. Mosquitoes were sorted and labeled according to locality and date, and stored in 1.5 ml Eppendorf tubes under dry conditions over silica gel until further analysis [13, 39]. Additionally, a laboratory strain of An. maculatus originating from Kulon Progo and under continuous colonization for greater than 30 years at the NIHRD-IVRCRD laboratory in Salatiga, Central Java [23] was compared with more recent field samples from Kulon Progo collected in 2015. To maintain the colony established at IVRCRD Salatiga, wild type material was re-introduced into the laboratory colony in 2003. This re-introduced wild type material was collected in the exact same location as the initial population, i.e. the village of Hargotirto, Kokap subdistrict, Kulon Progo district, Province of Yogyakarta. Representative field-collected specimens are deposited in the Systematics and Reference Laboratory, IVRCRD, Salatiga.

Fig. 1
figure 1

Map of the sampling sites in Indonesia. Each sampling site is indicated as a black spot. The name of sampling sites and samples are indicated. The source of geographical data layers is the Indonesia Geospatial Information Agency who granted the permission and rights to publish this map

Full size image
Table 1 Sampling localities and specimens of Anopheles mosquitoes
Full size table

DNA extraction, amplification and sequencing

DNA was extracted from the legs of each mosquito using a DNeasy® Blood & Tissue Kit (Qiagen, Hilden, Germany) with modification based on the manufacturer’s protocol. The amplification of ITS2 was performed with primers ITS2a (5′-TGT GAA CTG CAG GAC ACA T-3′) and ITS2b (5′-TAT GCT TAA ATT CAG GGG GT-3′) [39]. cox1 was amplified using the primers CI-N-2087 (5′-AAT TTC GGT CAG TTA ATA ATA TAG-3′) and TY-J-1460 (5′-TAC AAT TTA TCG CCT AAA CTT CAG CC-3′). PCR reactions were carried out using GoTaq® Green Master Mix (Promega, Madison, WI, USA). PCR thermocycling conditions for ITS2 were as follows: 94 °C for 10 min; followed by 40 cycles of denaturation at 94 °C for 1 min, annealing at 56 °C for 45 s and elongation at 72 °C for 1 min; followed by a final extention step at 72 °C for 10 min. For amplification of the cox1 gene, the following conditions were used: initial denaturation at 94 °C for 1 min followed by five cycles of 94 °C for 30 s, 45 °C for 40 s and 72 °C for 1 min; this was then followed by 35 cycles of 94 °C for 30 s, 55 °C for 40 s and 72 °C for 1 min, and by a final extention step at 72 °C for 10 min [40]. The amplified PCR products were separated by 1.5% agarose gel electrophoresis and vizualized by SYBR® safe DNA gel stain (Invitrogen, Carlsbad, CA, USA). A 100-bp DNA ladder was used for calculating the size of the PCR products. Amplification products were purified using Applied Biosystems ExoSAP-IT™ (Thermo Fisher Scientific, Vilnius, Lithuania). Cycle sequencing was performed using the primers listed above and an Applied Biosystems BigDye™ Terminator v.3.1 Cycle Sequencing Kit (Life Technologies Cooperation, Austin, TX, USA). To remove unicorporated BigDye® terminators and salts, cycle sequencing products were purified using a BigDye® Xterminator Purification Kit (Life technologies, Bedford, MA, USA). Sequence data were obtained using a DNA sequencer (Applied Biosystems® 3500 Genetic Analyzer) and analyzed using the Sequencing Analysis 6 program (Applied Biosystems).

Sequence analysis

Sequences were edited using Sequencing Analysis v.5.2 (Applied Biosystems). Sequences were aligned with MUSCLE using SeaView v.4.7 [41] and Mega X [42]. Phylogenetic trees were constructed with the maximum likelihood (ML) method and the Kimura-2 (K80) evolutionary model in Mega X. To assess the ML tree reliability, bootstraps were tested with 1000 replicates. To estimate the evolutionary divergence between sequences, genetic distances were analyzed by pairwise distance (p-distance) methods [43] in Mega X. Divergence time was calculated based on previously reported estimates giving 1 million years (Myr) for 2.3% difference [44, 45]. Sequences are deposited in GenBank under the following accession numbers: ITS2: N2 (MK659798), N44 (MK659797), S9 (MK659795), S33 (MK659793), NT64 (MK659796), NT101 (MK659794), KP10 (MK659792), KP72 (MK659780), 1x (MK659773), 2M (MK675654), 4M (MK675653), P1 (MK656100), C1 (MK656095) and C2 (MK656096); cox1: N2 (MK683469), N44 (MK683470), S9 (MK683476), S33 (MK683477), NT64 (MK683473), NT101 (MK683474), KP10 (MK683471), KP72 (MK683472), 1x (MK683464), 2M (MK683465), 4M (MK683466), P1 (MK683475), C1 (MK683467) and C2 (MK683468).

Results

ITS2 diversity and phylogeny of Anopheles maculatus

The comparative analysis of the ITS2 sequences of all An. maculatus samples and of available reference sequences from other members of the Maculatus Group and select other Anopheles species present in Indonesia indicates that two populations of An. maculatus are present in Indonesia. Samples of An. maculatus coming from Purbalingga (P1), Cilacap (C1, C2), Belu (NT64, NT101), Sebatik Island (N2, N44) and Ogan Komering Ulu (S9, S33) displayed 100% genetic similarity and were also 100% identical to An. maculatus sequences from the mainland Asian continent (Fig. 2, Additional file 1: Table S1). The GenBank An. maculatus sequences used as reference corresponded to mosquitoes isolated in India (JQ446438), Thailand (DQ518615), Vietnam (AY803351), Malaysia (DQ518619), Cambodia (DQ518618) and China (DQ518616). For the nuclear ribosomal ITS2 sequence, the similarity between all An. maculatus reference sequences and sequences from samples P1, C1, C2, NT64, NT101, N2, N44, S9 and S33 indicated a high conservation and genetic homogeneity regardless of distribution and geographical distance (Fig. 2). There was also no difference between samples from the continental Asian land mass and island groups. Conversely, the samples KP10, KP72, 1x, 2M and 4M isolated from Kulon Progo did not cluster with the continental An. maculatus sequences producing a separate, genetically distinct and homogeneous group more closely related to An. dispar. The sequences 1x, 2M and 4M, which correspond to a laboratory strain of An. maculatus collected decades ago in Kulon Progo, were identical to those samples collected for this work, i.e. KP10 and KP72, indicating strong genetic stability after years of continuous colonization.

Fig. 2
figure 2

Phylogenetic analysis of the ITS2 sequences. Kulon Progo samples are identified with blue stars. The phylogenetic tree was constructed using the maximum likelihood (ML) method and the Kimura-2 (K80) evolutionary model in Mega X. To assess the ML tree reliability, bootstraps were tested at 1000 replicates

Full size image

This phylogenetic analysis indicates that although separated into two different genetic aggregates (groups), collectively the Indonesian samples belong to the Maculatus Group. With respect to genetic distances, the samples displayed two ranges of distances depending on the group considered (Additional file 1: Table S1). The Indonesian An. maculatus group, i.e. samples P1, C1, C2, NT64, NT101, S9, S33, N2 and N44, showed no genetic distance with the continental An. maculatus reference sequences. The ITS2 sequence appears entirely conserved. The genetic distance within the Maculatus Group ranged between 2.7–20.8%, while the genetic distance of the An. maculatus sequences from other group members ranged between 5–16.5% (Additional file 1: Table S1). The Kulon Progo samples (KP10, KP72, 1x, 2M and 4M) displayed no (0%) internal group distance and a 5.5–5.8% distance with the other An. maculatus sequences. Comparison with other members, the Kulon Progo sequences displayed distances of 3%, 6.5%, 7.7–8, 8%, 10.4–10.9%, 15.5–17% and 10.9% with An. dispar, An. greeni, An. dravidicus, An. sawadwongporni, An. willmori, An. pseudowillmori and An. rampae, respectively (Additional file 1: Table S1). The alignment of the Kulon Progo ITS2 sequences with An. maculatus (s.s.) from Indonesia, An. maculatus (s.s.) from mainland Asia, and An. dispar is provided in Additional file 2: Figure S1.

cox1 diversity and phylogeny of Anopheles maculatus

The comparative analysis of the mitochondrial cox1 sequences, indicative of the maternal lineage, showed that all the samples and reference sequences belonged to four genetically distinct and separated lineages, Lineage 1 being separated into two sublineages (Fig. 3a). Lineage 4 comprised only Anopheles sinensis (subgenus Anopheles, Hyrcanus Group) and was therefore used as outgroup for the rooted tree (Fig. 3b). Lineage 1a comprised Anopheles stephensi, Anopheles flavirostris, An. dispar (a Maculatus Group member), all the Indonesian samples including the Kulon Progo population, An. gambiae and Anopheles barbirostris (Clade I); while Lineage 1b included three out four species of the Leucosphyrus Complex, along with Anopheles macarthuri, a member of the Riparis Subgroup, and Anopheles cracens, a member of the Dirus Complex, all five belonging to the Leucosphyrus Group [14]. Lineage 2 comprised only An. balabacensis, the fourth species of the Leucosphyrus Complex. Lineage 3 comprised all of the An. maculatus reference samples and members of the Maculatus Group available in GenBank, excluding An. dispar and the Indonesian An. maculatus samples. Lineage 3 and Lineage 4 branched directly on the root; whereas, Lineage 1a and Lineage 1b were further separated by a bootstrap of 99. The Indonesian An. maculatus sequences within Lineage 1a grouping displayed some internal genetic variability. All Kulon Progo samples examined were identical, while genetic distances of up to 2.6% were observed with An. maculatus (s.s.) from Indonesian (Additional file 3: Table S2). With respect to the other members of Lineage 1a, the closest species was An. dispar with a percentage of divergence ranging between 7.8–8.4% depending on the sample. The divergence of the Indonesian samples with An. flavirostris (Minimus Subgroup) ranged between 12.2–13.5%, while An. stephensi (Neocellia Series) ranged between 10.4–12.2% (Additional file 3: Table S2). The alignment of the cox1 sequences of the Kulon Progo samples, An. maculatus (s.s.) from Indonesia, An. maculatus (s.s.) from mainland Asia, and An. dispar is provided in Additional file 4: Figure S2.

Fig. 3
figure 3

Phylogenetic analysis of the cox1 sequences. a Midpoint rooted tree. b Tree rooted using Anopheles sinensis as outgroup. Kulon Progo samples are identified with stars. Phylogenetic trees were constructed with the maximum likelihood (ML) method and the Kimura-2 (K80) evolutionary model in Mega X. To assess the ML tree reliability, bootstraps were tested with 1000 replicates

Full size image

Time of divergence

The time of divergence was calculated based on previously reported estimates of the variation of the cox1 gene in the genus Anopheles where 2.3% of divergence is estimated to correspond to 1 million years (Myrs) [42, 43]. The estimated time of divergence of An. maculatus (s.s.) from the Kulon Progo population and from An. dispar was estimated at between 26–26.2 Myrs, and between 30.2–30.9 Myrs, respectively, which corresponds to the Oligocene Epoch. The divergence of the Kulon Progo population from the other members of the Kulon Progo maternal lineage, i.e. An. dispar, An. flavirostris and An. stephensi, is dated 3.4 (Pliocene Epoch), 5.3 (Miocene Epoch) and 5.04 Myrs (Pliocene), respectively. The other Indonesian An. maculatus sequences displayed the same separation time with An. dispar as the Kulon Progo population with the exception of NT101 (Belu, East Nusa Tenggara), C1 (Cilacap, Central Java) and P1 (Purbalingga, Central Java), indicating separation around 3.13 and 3.65 Myrs ago (Pliocene), respectively. The separation of the Kulon Progo population from the other Indonesian An. maculatus samples was dated 0.65 (NT64, S9, S33), 0.43 (NT101, C1, P1) and 0.22 Myrs (C2, N2, N44), all corresponding to the latter part of the Pleistocene Epoch.

Discussion

Anopheles maculatus (s.s.) was previously believed to be the only member of the Maculatus Group present in Indonesia, a species regarded as an important vector of malaria in certain localities [6]. This Asian group is a diverse assemblage with at least nine described species, five of which fall into two subgroups [3, 10, 12]. Investigating the diversity of An. maculatus in Indonesia was therefore a prerequisite for a better understanding of the distribution, bionomics and variations in vector capacity over its wide geographical range. The investigation reported herein provides several conclusions. First, there is definitive evidence of at least two species within the Maculatus Group in Indonesia, confirming a recent report by Ali et al. [24], which raises the number of species to ten (none of which are nominal taxon) within the Maculatus Group. Secondly, An. maculatus (s.s.) appears genetically homogeneous throughout its geographical range in Asia. Thirdly, members of the group in Indonesia differ by maternal origin from all other members, including An. maculatus (s.s.) from mainland Asia, with the lone exception of An. dispar.

The Kulon Progo population has been suspected as a distinct species within the Maculatus Group based on unpublished work spanning several decades (MJB, personal comm). This population was recently proposed as a different species based on selected morphological characters and genetic (ITS2 and cox2) sequences [24]. The ITS2 phylogenetic analysis in the present study confirmed that the Kulon Progo population and all other known An. maculatus sequences analyzed in Indonesia are members of the Maculatus Group, yet they also comprise genetically distinct groups. ITS2 is not considered a good intraspecific marker due to its low evolution rate and high conservation; however, it is a good marker at the species level showing clear discrimination indicative of species separation [46,47,48,49,50]. The phylogenetic distance between the ITS2 sequences of the Kulon Progo population and other An. maculatus sample sequences both Indonesia and mainland Asia included in the analysis ranged between 5.5–5.8%. This is greater than the ITS2 distances displayed by other groups of closely related Anopheles species. For example, two sibling species in the Anopheles farauti complex (an assemblage of 8 species) differ by only 4.0% [51], while An. greeni and An. dispar (Maculatus Group) also differ by 4.0% [52]. In Africa, five species within the An. gambiae complex show intraspecific differences ranging between 0.4–1.6% [53], while two members of the An. dirus complex, An. dirus (s.s.) (formerly species A) and An. baimaii (species D), display 5.4% genetic distance [54]. The Kulon Progo population was shown to be morphologically distinct from An. maculatus (s.s.) while cross-mating experiments generated partially sterile hybrids [24]. The combined evidence confirms that the Kulon Progo population is a distinct species and one that likely extends throughout the greater Menoreh Hill region in central Java. Until a formal morphological description can be made, it is hereby proposed that the Kulon Progo population be designated an infrasubspecific entity, An. maculatus var. menoreh, in reference to its region of origin and to distinguish it from An. maculatus (s.s.).

The two Indonesian members of the Maculatus Group and An. dispar belong to the same mitochondrial lineage and differ from that of all other known members of the group outside Indonesia. Collectively, these data demonstrate the occurrence in Indonesia of an introgression of the An. maculatus (s.s.) chromosomal genome from continental to insular populations. A similar phenomenon of introgression has been demonstrated for Anopheles sundaicus complex in Southeast Asia [44]. Introgression is a key adaptive mechanism of Anopheles mosquitoes to exist in various environments [44, 55], and well described in the Anopheles gambiae complex [56, 57].

The Pleistocene Epoch (2.58 Myrs to 11,700 years ago) is believed to have played a key role in the distribution of Anopheles mosquitoes in Southeast Asia [46, 58,59,60]. The period was characterized by a series of glaciation and inter-glaciation periods, which generated dramatic climatic changes and large variations in sea level [61, 62]. During glaciation periods, islands west of the Wallace’s Line were interconnected on the same land mass known as the Sunda Shelf [63,64,65]; whereas, during inter-glaciation events the rainforest environments expanded, thus providing more favorable habitats for Anopheles mosquitoes while island landmasses remained isolated. Sulawesi, the eastern Lesser Sunda and Maluku island chains, and western New Guinea Island were isolated and separated by sea from the western half of the Indonesian archipelago, while the Philippines followed a different biogeographical evolution. Palawan Island (western Philippines) was then connected to the Sunda Shelf but later separated and collided with the mobile belt of the Philippine archipelago. This geological history induced by shifts in climate is considered to have greatly influenced the current structural diversity of Anopheles populations in Southeast Asia and the evolution of present-day species complexes through successive genetic expansions and bottlenecks [66].

Based on the molecular evidence, the Kulon Progo population appears more closely related to An. dispar, a species that appears confined to the northern Philippines. The most parsimonious way to explain this geographical discrepancy is that their common ancestors gradually moved from continental Asia to the current island territories during the Oligocene, which corresponds to the calculated separation of the Kulon Progo lineage from the continental An. maculatus (s.s.) lineage (between 23 and 26.4 Myrs). A movement of An. maculatus from the continent appears to have occurred before 3.4 Myrs ago (between the late Oligocene and early Pliocene epochs), the calculated time of separation between An. dispar and the Kulon Progo population. This event led to introgression of the An. maculatus chromosomal genome into at least a portion of the maternal lineage identified as Lineage 1. During the Pliocene (3.4 Myrs ago), the ancestor of An. dispar was separated from the main introgressed population, likely the result of the tectonic shift of Palawan Island towards the current Philippine archipelago. During the Pleistocene, increased volcanism occurred in central and eastern Java but the Kulon Progo area was naturally spared from the surrounding destruction and served as a relic forest refuge [67]. This isolation event occurred between 0.22 and 0.65 Myrs, which corresponds to the calculated separation time between the Kulon Progo population and the other Indonesian archipelagic An. maculatus populations. During the late Pleistocene period (200,000 to 11,700 years ago), at least one other species invasion and introgression by continental An. maculatus appears to have occurred which generated the current Indonesian populations of An. maculatus (s.s.). This timescale is in agreement with that calculated for the movements of populations and introgression detected in An. sundaicus in Southeast Asia [44, 45].

Conclusions

Anopheles maculatus, along with Anopheles balabacensis, is the main malaria vector species occuring in the Kulon Progo area and the greater Menoreh region [33,34,35,36,37, 68,69,70,71]. The evidence presented here confirms that the Kulon Progo population is a distinct species and one that likely extends throughout the greater Menoreh Hill region in central Java. There are now two recognized members of the Maculatus Group present in Indonesia. However, a detailed morphological description of this novel species is required to establish a new nominal taxon. To distinguish it from An. maculatus (s.s.), in the interim it is hereby proposed an infrasubspecific entity (‘variety’), An. maculatus var. menoreh.