http://orcid.org/0000-0003-0726-6146
Figures
Abstract
Anopheles balabacensis, the primary vector of Plasmodium knowlesi in Sabah, Malaysia, is both zoophilic and anthropophilic, feeding on macaques as well as humans. It is the dominant Anopheles species found in Kudat Division where it is responsible for all the cases of P. knowlesi. However there is a paucity of basic biological and ecological information on this vector. We investigated the genetic variation of this species using the sequences of cox1 (1,383 bp) and cox2 (685 bp) to gain an insight into the population genetics and inter-population gene flow in Sabah. A total of 71 An. balabacensis were collected from seven districts constituting 14 subpopulations. A total of 17, 10 and 25 haplotypes were detected in the subpopulations respectively using the cox1, cox2 and the combined sequence. Some of the haplotypes were common among the subpopulations due to gene flow occurring between them. AMOVA showed that the genetic variation was high within subpopulations as compared to between subpopulations. Mantel test results showed that the variation between subpopulations was not due to the geographical distance between them. Furthermore, Tajima’s D and Fu’s Fs tests showed that An. balabacensis in Sabah is experiencing population expansion and growth. High gene flow between the subpopulations was indicated by the low genetic distance and high gene diversity in the cox1, cox2 and the combined sequence. However the population at Lipasu Lama appeared to be isolated possibly due to its higher altitude at 873 m above sea level.
Citation: Manin BO, Drakeley CJ, Chua TH (2018) Mitochondrial variation in subpopulations of Anopheles balabacensis Baisas in Sabah, Malaysia (Diptera: Culicidae). PLoS ONE 13(8): e0202905. https://doi.org/10.1371/journal.pone.0202905
Editor: Tzen-Yuh Chiang, National Cheng Kung University, TAIWAN
Received: January 4, 2018; Accepted: August 10, 2018; Published: August 23, 2018
Copyright: © 2018 Manin et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All gene sequence files are available from the National Center for Biotechnology Information (NCBI) database with accession number starting from MH032606 to MH032747.
Funding: This study was supported by: Medical Research Council, Natural Environment Research Council, Economic and Social Research Council, and Biotechnology and Biosciences Research Council for the funding received for this project through the Environmental & Social Ecology of Human Infectious Diseases Initiative (ESEI), MRC Grant Number: G1100796. BOM would also like to thank the Malaysian Ministry of Higher Education or providing a scholarship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Anopheles spp. are the only vectors of human and zoonotic malaria caused by five malaria parasite species namely Plasmodium falciparum, P. vivax, P. malariae, P. ovale and P. knowlesi. Approximately 70 Anopheles species have been known to transmit these malaria parasites in nature and 41 of them are considered as dominant vector species/species complex [1–2]. Nineteen of them are found in Asia [2] with four species viz. Anopheles dirus, An. balabacensis, An. latens and An. introlatus belonging to the Leucosphyrus group [3–4]. Anopheles dirus, a member of the Dirus complex found mainly in China, Cambodia, Vietnam, Laos and Thailand is the primary vector for human and simian malaria in Vietnam [5]. In Malaysia, An. balabacensis, An. latens and An. introlatus, all members of the Leucosphyrus complex have been incriminated as primary vectors for P. knowlesi [6–8].
Anopheles balabacensis is found in the forested areas of Philippines (Balabac and Palawan), Indonesia (Kalimantan, Lombok, Java and Sumba), East Malaysia (Sabah and Sarawak) and Brunei [3–4, 9]. Recent studies conducted in Sabah showed that An. balabacensis prefers to bite humans outdoors rather than indoors [10] and during the early evening with the peak biting period between 7–8 pm [8, 11–12].
Sabah has the highest incidence of P. knowlesi malaria in the world with most of the cases reported in 2013 occurring in the interior areas [13]. Records of Sabah Department of Health show that the proportion of P. knowlesi among the indigenous malaria cases for 2014–2016 was respectively 66%, 80% and 92%. Asymptomatic infection has also been detected in the community. A survey conducted in Kudat and Kota Marudu districts found that 9.8% (112/1147) of the collected blood samples were positive for P. knowlesi with the majority of the infected individuals not having a history of fever [14]. In Sabah P. knowlesi has caused most of the malaria deaths in adults [15]. Although An. balabacensis has been confirmed as the main vector for both P. falciparum [10] and P. knowlesi [8] population genetics of An. balabacensis in Sabah not been investigated.
We conducted a study on the genetic variation between subpopulations of An. balabacensis in Sabah based on the cox1, cox2 and the combined cox1 and cox2 sequences (“combined sequence”) of mitochondrial DNA. The mitochondrial DNA was used in the study as it is a suitable marker in a wide range of taxonomic, population and evolutionary studies in animals including malaria vectors [16–18]. Such population genetic analysis will help in understanding the evolution and gene flow of An. balabacensis populations.
Materials and methods
Collection sites
All the study sites selected had previous records of P. knowlesi cases. The inter-site distance varied from 2.4 km to 237.2 km with the GPS coordinates varying from 5.33192N - 7.21578N to 116.04140E - 117.10292E (Fig 1). The greatest inter-site distance of 237.2 km was between Limbuak Laut in Banggi Island and Keritan Ulu in mountainous Keningau district, while the shortest distance of 2.4 km was between Tinukadan Laut and Membatu Laut, both in Kudat district. The subpopulations at Limbuak Laut and Timbang Dayang are located in Banggi Island while the rest are in the main Borneo Island. The subpopulations located at the northern part of Sabah (e.g. Sorinsim, Lipasu Lama and Paus) however are separated from the Keritan Ulu subpopulation by Crocker Range and Mount Trus Madi.
There were 14 sampling sites, each denoted by a different number. Seven sites were located in Kudat district, two in Banggi Island, and one each at the other five study sites. The outline of the map and the elevation map were downloaded from open source websites: http://gadm.org/country and http://www.diva-gis.org/gdata respectively, the final map was created using QGIS software version 2.18.13.
Mosquito collection and morphological identification
Anopheles specimens were collected from February, 2014 to September, 2016 using human landing catch method (HLC) (S1 Table). Each mosquito was kept separately inside a tube with collection details on the locality and the time caught. Any specimen still alive the next day would be killed by keeping it in the freezer (-20°C) for 3–5 minutes. The specimens were identified to species level using Anopheles identification keys [3,19–20] and An. balabacensis were isolated and kept individually each in a 1.5 ml microfuge tube at -30°C until use.
PCR amplification and sequencing of the cox1 and cox2 mitochondrial gene fragments
Genomic DNA was extracted from each An. balabacensis using the DTAB-CTAB method [21] and stored at -30°C until use. Nested PCR was performed to amplify the cox1 and cox2 genes. Details of the PCR primers used are shown in S2 Table and the binding sites of the PCR primers illustrated in S1 Fig. The PCR mixture was prepared from PCR kit (Promega, USA) by mixing 10.0 μl of 5X PCR buffer, 1.0 μl of dNTPs (10 mM), 5.0 μl of MgCl2 (25 mM), 2.0 μl of the forward and reverse primers (10 μM), 1.0 μl of Taq DNA polymerase (5.0 U/μl), 3.0 μl of DNA template and 26.0 μl sterile dH2O. After the first PCR reaction was completed, 3.0 μl of the PCR product was used as a DNA template in the second PCR. The PCR reaction was performed using a thermal cycler (T100 Thermal Cycler, BioRad) with an initial denaturation at 95°C for 5 min followed by 30 cycles of denaturation at 94°C for 1 min, annealing at 55°C for 1 min, extension at 72°C for 1 min and one final extension step at 72°C for 10 min.
After the PCR was completed, the PCR products were purified using MEGA quick-spin PCR & Agarose Gel DNA Extraction System (iNtRON Biotechnology, Korea) according to the manufacturer’s procedure. The purified PCR products were analyzed on 1.5% agarose gel electrophoresis stained with RedSafe nucleic acid staining solution (iNtRON Biotechnology) and visualized using UV transilluminator. The purified PCR products were sent to AITBIOTECH (Singapore) for sequencing using forward and reverse primers (cox1—COIF+UAE10; cox2—X2F+COIIR). In order to determine the consistency of the Taq DNA polymerase, eleven PCR products of An. balabacensis cox1 and cox2 genes from Paradason were cloned into pGEM-TEasy vectors (Promega, USA) and the plasmids were extracted from the transformed E. coli (JM109) using DNA-spin Plasmid DNA Purification Kit (iNtRON Biotechnology, Korea) all according to the manufacturer’s procedure. The extracted plasmid was restricted using EcoRI restriction enzyme (Promega, USA) and the two plasmids from each gene containing the correct size of PCR amplicon were sent to AITBIOTECH, Singapore for sequencing at both directions using forward and reverse M13 primers.
Data analysis
Cox1 and cox2 genes of 71 An. balabacensis individuals were sequenced. These sequences have been uploaded in the National Center for Biotechnology Information (NCBI) database with accession number starting from MH032606 to MH032747 (S3 Table). Subsequent analyses were performed separately using cox1, cox2 and the combined sequence.
The sequences were multi-aligned using ClustalW incorporated in the MEGA4.1 software [22]. The nucleotide sequences selected for alignment were respectively nt1509—nt2891 (1,383 bp) for cox1 and nt3029—nt3713 (685 bp) for cox2 with reference to the nucleotide sequence of An. cracens (JX219733) [23]. Cox1, cox2 and the combined sequence were translated into proteins based on the genetic code for mitochondrial DNA of Drosophila using DnaSP software (ver. 5.10.01) [24]. The population structure of An. balabacensis subpopulations was explored with molecular variance analysis (AMOVA) using Arlequin 3.11 [25]. The population pairwise FST values for genetic distance between the subpopulations were tested for significance as well as used for estimating gene flow, using 1,000 permutations [26]. The number of haplotypes in the subpopulations, the average number of nucleotide differences, haplotype diversity [27] and nucleotide diversity [28] were also estimated using Arlequin 3.11.
Neutrality test using Tajima’s D [29] and Fu’s Fs [30] was carried out with 1,000 simulations to analyse the randomness of the DNA sequence evolution. We further investigated the demographic expansion with mismatch analysis test using the sum of squared deviation values (SSD) and raggedness index (Rag). Estimation of the time interval for the population expansion was done using the expression, [31], where τ is the estimated number of generations since the expansion,
the mutation rate per site per generation, and
the sequence length. A mutation rate of 1.15 x 10−8 [32] was used.
Mantel test for isolation by distance (IBD) was performed online (http://ibdws.sdsu.edu/) with 10,000 permutations to assess the significance of correlation between genetic distance and linear geographical distance [33]. The test was conducted first, for all the subpopulations and subsequently, for the subpopulations on the main island only. The haplotype network was estimated and drawn using statistical parsimony method [34] incorporated in PopART software (http://popart.otago.ac.nz).
Ethical clearance
This study was approved by the National Medical Research Register of the Malaysian Ministry of Health (NMRR, Ref.NMRR-12-786-13048). Consent to carry out mosquito collection was obtained from the village council or the village headman and the land owners. All volunteers who carried out mosquito collections signed informed consent forms and were provided with antimalarial prophylaxis during the study period.
Results
Cox1 and cox2 sequences of An. balabacensis
The cox1 sequence had 31.1% A, 38.4% T, 15.8% C and 14.7% G with an A + T bias of 69.5%, and can be translated into 461 amino acids. There were more mutations by transition (93.65%) than transversion (6.35%). In the transition mutations-, inter-changes between the two-ring purines: A → G (22.17%); G → A (47.03%) were more frequent than one-ring pyrimidines: C → T (17.31%); T → C (7.14%).
The cox2 sequence had 35.9% A, 38.4% T, 13.4% C and 12.3% G with an A + T bias of 74.3% and can be translated into 228 amino acids. Mutation by transition (86.89%) was more common than by transversion (13.11%). However in the transition, only inter-changes between two-ring purines: A → G (22.11%); G → A (64.78%) were detected.
Mitochondrial diversity
Based on cox1, the subpopulations have the following genetic statistics: number of haplotypes 1–5, haplotype diversity 0–1, nucleotide diversity 0–0.00231, average number of nucleotide differences 0–3.2 and number of segregating sites 0–8 (Table 1). The subpopulation of Mambatu Laut had the highest haplotype diversity while Tomohon had the highest nucleotide diversity.
Based on cox2, the subpopulations have the following genetic statistics: number of haplotypes 1–3, haplotype diversity 0–0.833, nucleotide diversity 0–0.00195, average number of nucleotide differences 0–1.333 and number of segregating sites 0–2 (Table 2). Sinangip subpopulation had the highest haplotype diversity while Lipasu Lama had the highest nucleotide diversity.
For the combined sequence, the genetic statistics were as follows: number of haplotypes 1–5, haplotype diversity 0–1, nucleotide diversity 0 to 0.00193, average number of nucleotide differences 0–4 and number of segregating sites 0–10 (Table 3). The subpopulations of Mambatu Laut and Sinangip had the highest haplotype diversity while Tomohon had the highest nucleotide diversity.
Haplotype diversity
Based on cox1, a total of 17 haplotypes were detected from the subpopulations (Fig 2A), with Hap_1 having the highest frequency (n = 27, 38.0%) followed by Hap_2 (n = 16, 22.5%) and Hap_6 (n = 7, 9.9%). Six haplotypes (Hap_1, Hap_2, Hap_3, Hap_6, Hap_7 and Hap_10) were shared in at least two subpopulations (S4 Table). Hap_1 was found in 12 subpopulations except in Sorinsim and Lipasu Lama, while Hap_2 was found in nine subpopulations except in Tomohon, Minikodong, Timbang Dayang, Sinangip and Lipasu Lama. Hap_6 was found in five subpopulations (Tinukadan Laut, Mambatu Laut, Minikodong, Sinangip and Lipasu Lama). Eleven haplotypes were unique, two each in Paradason, Tomohon and Timbang Dayang, but one each in Mambatu Laut, Narandang, Limbuak Laut, Lipasu Lama and Paus.
Frequency of haplotype detected in (a) cox1, (b) cox2 and (c) combined sequence across all the subpopulations of An. balabacensis.
As for cox2, a total of ten haplotypes were observed from the subpopulations (Fig 2B). Hap_1 had the highest frequency (n = 50, 70.4%), followed by Hap_4 (n = 7, 9.9%). Five haplotypes (Hap_1, Hap_3, Hap_4, Hap_5 and Hap_6) were shared at least in two subpopulations (S4 Table). Hap_1 was found in all the subpopulations, whereas Hap_4 was found in Tomohon, Timbang Dayang, Limbuak Laut and Sinangip. Five haplotypes were unique, two of them recorded in Mambatu Laut, one each in Paradason, Lipasu Lama and Paus.
For the combined sequence, a total of 25 haplotypes were obtained (Fig 2C). Hap_1 had the highest frequency (n = 17, 23.9%), followed by Hap_2 (n = 15, 21.1%). Seven haplotypes (Hap_1, Hap_2, Hap_6, Hap_7, Hap_8, Hap_9, and Hap_13) were shared in at least two subpopulations (S4 Table). Hap_1 was found in seven subpopulations, mainly in Kudat and Keningau districts, while Hap_2 was detected in nine subpopulations except in Tomohon, Minikodong, Timbang Dayang, Sinangip and Lipasu Lama. Eighteen haplotypes were unique, three each in Paradason and Timbang Dayang, two each in Mambatu Laut, Tomohon, Lipasu Lama and Paus, and one each in Longgom Besar, Narandang, Limbuak Laut and Keritan Ulu.
Population genetic structure
In the hierarchical AMOVA, the variance component for cox1 was higher within subpopulations than among subpopulations (89.84% vs 10.16%; p<0.05, Table 4). Sorinsim subpopulation had the highest FST value (0.282), while Timbang Dayang the lowest (-0.082) (S5 Table). The pairwise FST values range from -0.333 (between Longgom Besar and Keritan Ulu) to 0.600 (between Lipasu Lama and Sorinsim) with gene flow among the subpopulations varying from 0.333 to ∞ (S6 Table).
As for cox2, the variance component was also higher within subpopulations compared to among subpopulations (84.70% vs 15.30%; p<0.05, Table 4). Tinukadan Laut and Sorinsim subpopulations had the highest FST value (0.323) while Lipasu Lama the lowest (-0.014) (S5 Table). The pairwise FST values range from -0.333 (between Longgom Besar and Keritan Ulu) to 0.634 (between Lipasu Lama and Tinukadan Laut) with gene flow among the subpopulations varying from 0.289 to ∞ (S7 Table).
Similarly the combined sequence also had higher variance component for within subpopulations compared to among subpopulations (87.06% vs 12.94%; p<0.05, Table 4). Sorinsim subpopulation had the highest FST value (0.304), while Mambatu Laut the lowest (0.082) (S5 Table). The pairwise FST values range from -0.212 (between Narandang and Keritan Ulu) to 0.667 (between Lipasu Lama and Sorinsim and Minikodong and Sorinsim) with gene flow among the subpopulations varying from 0.25 to ∞ (S8 Table).
The neutrality test showed four subpopulations (Paradason, Tinukadan Laut, Mambatu Laut and Sinangip) had either significant Tajima’s D (p<0.01) or Fu’s Fs (p<0.001) values for either cox gene or for the combined sequence. Sorinsim subpopulation had zero values for Tajima’s D and Fu’s Fs for cox1, cox2 and the combined sequence (Table 5).
Mismatch analysis showed that, the overall SSD value and the Rag index were not significant for both cox genes (cox1—SSD = 0.001, p = 0.849; Rag = 0.027, p = 0.814; cox2—SSD = 0.000, p = 0.8700; Rag = 0.092, p = 0.5440), but the SDD value was significant for the combined sequence (SSD = 0.044, p = 0.0060; Rag = 0.025, p = 0.962) (Table 6). At the subpopulation level, all except Tinukadan Laut and Sorinsim subpopulations show non-significant SSD values and Rag index for both genes. For the combined sequence, three subpopulations (Paradason, Lipasu Lama and Paus) showed significant SSD values, while Sorinsim subpopulation had significant SSD value and Rag index.
The graph of the mismatch distribution for cox1, cox2 and the combined sequence showed a unimodal peak indicating the population expansion model is applicable (Fig 3). Using the expected τ values (cox1: 0.992; cox2: 0.684; combined sequence: 1.438, where ) obtained from the expansion model, the expansion event was estimated to have taken place between 3,600 to 2,500 years ago, assuming one generation of An. balabacensis per month based on laboratory data.
Graphs of the mismatch distribution analysis for total populations of An. balabacensis based on (a) cox1, (b) cox2 and (c) the combined sequence. The dotted lines represent the observed frequency of pairwise differences, and the solid lines show the expected values for the population expansion model.
Mantel test for isolation by distance showed that the regression of the genetic distance (or linearized FST values = FST/(1-FST)) on geographical distance was not significant (Fig 4).
Plot of genetic distance against geographical distance between pairs of An. balabacensis subpopulations based on (a) cox1, (b) cox2 and (c) the combined sequence.
Genealogical relationship among haplotypes
The haplotype network shows that An. balabacensis of Sabah belongs to one cluster derived from a single ancestral haplotype. Based on cox1 (Fig 5A), Hap_2 is considered the ancestral haplotype which is connected to the other 16 haplotypes by 1–5 mutation steps. In cox2 haplotype network (Fig 5B), Hap_1, the dominant haplotype is considered to be the ancestral haplotype and is connected to the other 9 haplotypes by 1–2 mutation steps. For the combined sequence, Hap_1 considered the ancestral haplotype and is connected to 24 other haplotypes by 1–6 mutation steps (Fig 5C).
Genealogical relationship among the haplotypes based on (a) cox1, (b) cox2 and (c) combined sequence of An. balabacensis as estimated by statistical parsimony. Each haplotype is represented by a different number inside the circle. The size of a circle is proportional to the frequency of the haplotype. The hatch marks on the line represent mutations.
There are 23 mutation sites identified in cox1 sequence (Fig 6A), of which, 14 are synonymous while 9 are non-synonymous mutations (Fig 6B). All the synonymous mutations are located at the third codon, whereas 3 non-synonymous mutations are sited at the first codon, 5 at second codon and 1 at third codon.
Variable positions of (a) nucleotide and (b) amino acid in cox1 sequence of An. balabacensis. The numbers shown above the sequences represent the nucleotide or amino acid position and the dots refer to the identity with reference to ancestral haplotype (Hap_2).
For cox2, 11 mutations are recorded in the sequence (Fig 7A), 8 of which were synonymous while 3 non-synonymous at the first, second and third codons (Fig 7B).
Variable positions of (a) nucleotide and (b) amino acid in cox2 sequence of An. balabacensis. The numbers shown above the sequences represent the nucleotide or amino acid position and the dots refer to the identity with reference to ancestral haplotype (Hap_1).
As for the combined sequence, 34 mutations (22 synonymous and 12 non-synonymous) were detected (Fig 8).
Variable positions of (a) nucleotide and (b) amino acid in the combined sequence of An. balabacensis. The numbers shown above the sequences represent the nucleotide or amino acid position and the dots refer to the identity with reference to ancestral haplotype (Hap_1).
Discussion
The genetic variation of An. balabacensis populations in Sabah was explored by analyzing the partial sequence of cox1 (1,383 bp), cox2 (685 bp) and the combined sequence (2,068 bp) of the mitochondrial DNA from 71 specimens collected from 14 different sites each representing a different subpopulation.
Overall, the genetic distance between the subpopulations was low (AMOVA: FST value: cox1: 0.102; cox2: 0.153; combined sequence: 0.129), likely to be a result of inter-breeding and gene flow between the subpopulations.
Based on cox1 sequence the ancestral haplotype was found in five districts viz. Banggi (Limbuak Laut), Kudat (Paradason, Longgom Besar, Tinukadan Laut, Mambatu Laut, Narandang), Kota Marudu (Sorinsim), Ranau (Paus) and Keningau (Keritan Ulu) which are geographically far apart from each another, while based on the cox2 the ancestral was detected in all the subpopulations. However, the combined sequence showed that the ancestral was found only at two districts viz. Kudat (Paradason, Longgom Besar, Tinukadan Laut, Narandang, Tomohon and Minikodong) and Keningau (Keritan Ulu). It is unlikely that the total number of haplotypes in a subpopulation had been sampled in our study, for this would depend on the sample size [35]. However there is no obvious way to decide on required sample size based on traditional approaches [36]. The highest haplotype diversity of cox1 was observed in the Mambatu Laut subpopulation while the highest nucleotide diversity was in Tomohon subpopulation, both situated in the low lying areas of Kudat. Based on cox2 sequence, the Sinangip subpopulation has the highest haplotype diversity and Lipasu Lama has the highest nucleotide diversity, both located in higher altitude areas. The Sorinsim subpopulation had zero diversity for both sequences which could probably be due to sampling error.
Our results provide some basic information on the genetic structure of An. balabacensis especially those collected from the higher altitude areas. Cox1 has higher haplotype diversity and nucleotide diversity compared to cox2, which would suggest that cox1 is a more suitable molecular marker for investigating genetic variation and structure of An. balabacensis. It has been shown the number of samples and the area covered in such population study would influence the data obtained and thus the interpretation [36–37]. To obtain a better picture of the genetic variation of the population than either cox1 or cox2 alone, the complete mitochondrial DNA sequence would be required [38–39].
In general, our results show that the genetic diversity in the cox1, cox2 and the combined sequence among the An. balabacensis subpopulations was moderate to high. This indicates high gene flow between the subpopulations which may imply that An. balabacensis has high dispersal rate contributing to its success as a vector for P. knowlesi.
The sequences of both cox1 and cox2 of An. balabacensis contain a high ratio of A + T similar to other Anopheles species [40–42], with relatively much higher A + T at the third codon. Only mutations by transition and transversion have been detected in both sequences. The majority of these mutations are located at the third codon and higher mutation rate by transition is a common feature between the same or related species [16]. Similar finding was reported for other Anopheles spp.: An. oswaldoi in Brazil and for An. minimus in China and South East Asia [41–43]. However, mutations at the third codon usually do not result in altering of the amino acid composition because of the redundancy effect [43–44].
AMOVA of the sequences showed that the genetic variation in An. balabacensis subpopulations in Sabah lie within subpopulations rather than among subpopulations. Similar finding has also been recorded for other Anopheles species, e.g. An. baimaii (also of the Leuscosphyrus group) populations in India [45], An. lesteri [46] and An. sinensis [37,47] populations in China. In Yunnan, located in the mountainous area, the An. sinensis population was unique compared to other subpopulations in China [37]. It had been suggested that the physical barrier and the heterogeneous landscape could have inhibited gene flow between Yunnan and the other subpopulations. However for An. balabacensis in this study, the evidence is not strong enough to substantiate the conclusion that any subpopulation is isolated geographically. Nevertheless the high pairwise genetic distance and the low gene flow for the Lipasu Lama subpopulation located in the mountainous area (873 meter above the sea level), may indicate possible early stage of isolation. Another subpopulation that may also show early stage of isolation is Sorinsim, as indicated by the combined sequence analysis. It is possible that the presence of cryptic species may also contribute to the observed high pairwise of genetic distance in these two subpopulations, but this needs to be confirmed by further study.
The Limbuak Laut and Timbang Dayang subpopulations located in Banggi Island which is separated from the main island by 44 km showed small to moderate pairwise genetic distance and high gene flow, indicating that there is also sharing of genetic material between Banggi and the main island subpopulations through breeding and migration. This could possibly be achieved by An. balabacensis adults being transported unintentionally in boats or ferry along with the daily movement of people between the main island and Banggi Island.
The negative Tajima’s D values obtained in the overall subpopulations suggest that the DNA sequences are evolving in a non-random manner and many rare alleles are present in the subpopulations which are expanding demographically [29]. This is supported by the strong significant negative values for Fu’s Fs which is a more sensitive statistic for detecting deviations from neutrality and indicators for population expansion and growth [30]. A single unimodal peak (Fig 3) and the small non-significant values of the mismatch distribution analysis further support the population expansion hypothesis [48]. Furthermore, Mantel testing did not show any isolation by distance in these subpopulations indicating that the genetic variation was not caused by the geographical distance. Similar results were also observed in An. dirus and An. baimaii in South East Asia [40], An. baimaii in north-east India [45], An. sinensis [37] and An. lesteri [46] both in China, showing there is an excess of rare alleles in these populations.
The genealogy networks of cox1, cox2 and the combined sequence showed that An. balabacensis population of Sabah belongs to one cluster, suggesting that the subpopulations are expanding from a single ancestral haplotype. Based on cox1, this ancestral haplotype (Hap_2) was found in nine subpopulations, whereas based on cox2, the ancestral haplotype (Hap_1) was found in all the subpopulations. However the combined sequence showed that the ancestral haplotype (Hap_1) was found in Kudat and Keningau districts. It appears that the mutation rate of cox1 and cox2 differ [49–50], resulting in different number of haplotypes, unique haplotypes and thus different haplotype network. However, the presence of the same haplotypes in different subpopulations suggests that inter-breeding and migrations might have been occurring between the subpopulations.
This study has shown that An. balabacensis population of Sabah is undergoing population growth and expansion. The low genetic distance in the overall population of An. balabacensis based on the mitochondrial DNA indicates that there is high genetic diversity in the subpopulations, a likely consequence of inter-population migration and breeding, resulting in gene flow between the subpopulations.
Supporting information
S1 Fig. The annealing direction of the primers for cox1 and cox2 genes of An. balabacensis during the PCR amplification.
https://doi.org/10.1371/journal.pone.0202905.s001
(TIF)
S1 Table. Details on the collection dates, sites and number of An. balabacensis collected in this study.
https://doi.org/10.1371/journal.pone.0202905.s002
(PDF)
S2 Table. PCR primers used to amplify cox1 and cox2 genes of An. balabacensis.
https://doi.org/10.1371/journal.pone.0202905.s003
(PDF)
S3 Table. GeneBank accession and haplotype numbers of the 71 An. balabacensis specimens.
https://doi.org/10.1371/journal.pone.0202905.s004
(PDF)
S4 Table. Number and frequency of haplotypes observed for cox1, cox2 and the combined sequences.
The haplotypes marked with asterisk (*) were detected only in one subpopulation.
https://doi.org/10.1371/journal.pone.0202905.s005
(PDF)
S5 Table. Fixation index (FST) among populations of An. balabacensis calculated based on the cox1, cox2 and the combined sequence.
https://doi.org/10.1371/journal.pone.0202905.s006
(PDF)
S6 Table. Pairwise genetic distance (FST) and gene flow (Nm) between subpopulations of An. balabacensis based on the cox1.
Nm values are shown above the diagonal while FST values below the diagonal. Values marked with asterisk indicate the genetic distances between two subpopulations are significant: *p<0.05, **p<0.01.
https://doi.org/10.1371/journal.pone.0202905.s007
(PDF)
S7 Table. Pairwise genetic distance (FST) and gene flow (Nm) between subpopulations of An. balabacensis based on the cox2.
Nm values are shown above the diagonal while FST values below the diagonal. Values marked with asterisk indicate the genetic distances between two subpopulations are significant: *p<0.05.
https://doi.org/10.1371/journal.pone.0202905.s008
(PDF)
S8 Table. Pairwise genetic distance (FST) and gene flow (Nm) between subpopulations of An. balabacensis based on the combined sequence.
Nm values are shown above the diagonal while FST values below the diagonal. Values marked with asterisk indicate the genetic distances between two subpopulations are significant: *p<0.05, **p<0.01, ***p<0.001.
https://doi.org/10.1371/journal.pone.0202905.s009
(PDF)
References
- 1. Hay SI, Sinka ME, Okara RM, Kabaria CW, Mbithi PM, Tago CC, et al. Developing global maps of the dominant Anopheles vectors of human malaria. PLoS Med. 2010;7(2):e1000209. pmid:20161718
- 2. Sinka ME, Bangs MJ, Manguin S, Rubio-Palis Y, Chareonviriyaphap T, Coetzee M, et al. A global map of dominant malaria vectors. Parasites & Vectors. 2012;5:69.
- 3. Sallum MAM, Peyton EL, Harrison BA, Wilkerson RC. Revision of the Leucosphyrus group of Anopheles (Cellia) (Diptera, Culicidae). Rev Bras entomol. 2005;49(1):1–152.
- 4. Vythilingam I, Wong ML, Wan-Yussof WS. Current status of Plasmodium knowlesi vectors: a public health concern? Parasitology. 2016 May 25:1–9.
- 5. Nakazawa S, Marchand RP, Quang NT, Culleton R, Manh ND, Maeno Y. Anopheles dirus co-infection with human and monkey malaria parasites in Vietnam. Int J Parasitol. 2009;39:1533–1537. pmid:19703460
- 6. Vythilingam I, Tan CH, Asmad M, Chan ST, Lee KS, Singh B. Natural transmission of Plasmodium knowlesi to humans by Anopheles latens in Sarawak, Malaysia. Trans R Soc Trop Med Hyg. 2006;100:1087–1088. pmid:16725166
- 7. Vythilingam I, Lim YA, Venugopalan B, Ngui R, Leong CS, Wong ML, et al. Plasmodium knowlesi malaria an emerging public health problem in Hulu Selangor, Selangor, Malaysia (2009–2013): epidemiologic and entomologic analysis. Parasites & Vectors. 2014;7:436.
- 8. Wong ML, Chua TH, Leong CS, Khaw LT, Fornace K, Wan-Sulaiman WY, et al. Seasonal and spatial dynamics of the primary vector of Plasmodium knowlesi within a major transmission focus in Sabah, Malaysia. PLoS Negl Trop Dis. 2015;9(10):e0004135. pmid:26448052
- 9. Sinka ME, Bangs MJ, Manguin S, Chareonviriyaphap T, Patil AP, Temperley WH, et al. The dominant Anopheles vectors of human malaria in the Asia-Pacific region: occurrence data, distribution maps and bionomic précis. Parasites & Vectors. 2011;4:89.
- 10. Hii JLK, Kan S, Vun YS, Chin KF, Tambakau S, Chan MKC, et al. Transmission dynamics and estimates of malaria vectorial capacity for Anopheles balabacensis and An. flavirostris (Diptera: Culicidae) on Banggi island, Sabah, Malaysia. Ann Trop Med Parasitol. 1988;82(1):91–101. pmid:3041932
- 11. Brant HL, Ewers RM, Vythilingam I, Drakeley C, Benedick S. Vertical stratification of adult mosquitoes (Diptera: Culicidae) within a tropical rainforest in Sabah, Malaysia. Malar J. 2016;15:370. pmid:27430261
- 12. Manin BO, Ferguson HM, Vythilingam I, Fornace K, William T, Torr SJ, et al. Investigating the contribution of peri-domestic transmission to risk of zoonotic malaria infection in humans. PLoS Negl Trop Dis. 2016;10(10):e0005064. pmid:27741235
- 13. William T, Jelip J, Menon MY, Anderios F, Mohammad R, Mohammad TAAM, et al. Changing epidemiology of malaria in Sabah, Malaysia: increasing incidence of Plasmodium knowlesi. Malar J. 2014;13:390. pmid:25272973
- 14. Fornace KM, Nuin NA, Betson M, Grigg MJ, William T, Anstey NM, et al. Asymptomatic and submicroscopic carriage of Plasmodium knowlesi malaria in household and community members of clinical cases in Sabah, Malaysia. J Infect Dis. 2016;213:784–787. pmid:26433222
- 15. Rajahram G, Barber BE, William T, Grigg MJ, Menon J, Yeo TW, et al. Falling Plasmodium knowlesi malaria death rate among adults despite rising incidence, Sabah, Malaysia, 2010–2014. Emerg Infect Dis. 2016;22(1):41–48. pmid:26690736
- 16. Moritz C, Dowling TE, Brown WM. Evolution of animal mitochondrial DNA: relevance for population biology and systematic. Ann Rev Ecol Syst. 1987;18:269–292.
- 17. Lunt DH, Zhang DX, Szymura JM, Hewitt GM. The insect cytochrome oxidase I gene: evolutionary patterns and conserved primers for phylogenetic studies. Insect Mol Biol. 1996;5(3):153–165. pmid:8799733
- 18. Norris DE. Genetic markers for study of the anopheline vectors of human malaria. Int J Parasitol. 2002;32:1607–1615. pmid:12435445
- 19. Cagampang-Ramos A, Darsie RF Jr. Illustrated keys to the Anopheles mosquitoes of the Philippine Islands. USAF Fifth Epidemiological Flight, PACAF, Technical Report 1970;(70):1–49.
- 20. Rattanarithikul R, Harrison B, Harbach RE, Panthusiri P, Coleman RE. Illustrated keys to the mosquitoes of Thailand IV. Anopheles. Southeast Asian J Trop Med Public Health. 2006;37:1–128.
- 21. Phillips A, Simon C. Simple, efficient, and nondestructive DNA extraction protocols for arthropods. Ann Entomol Soc Am. 1995;88(3):281–283.
- 22. Tamura K, Dudley J, Nei M, Kumar S. MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol Biol Evol. 2007;24:1596–1599. pmid:17488738
- 23. Logue K, Chan ER, Phipps T, Small ST, Reimer L, Henry-Halldin C, et al. Mitochondrial genome sequences reveal deep divergences among Anopheles punctulatus sibling species in Papua New Guinea. Malar J. 2013;12:64. pmid:23405960
- 24. Rozas J, Sanchez-Delbarrio J, Anchez-Delbarrio J, Messeguer X, Rozas R. Dnasp: DNA polymorphism analyses by the coalescent and other methods. Bioinformatics. 2003;19:2496–2497. pmid:14668244
- 25. Excoffier L, Laval G, Schneider S. Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evol Bioinform Online 2005, 1:47–50.
- 26. Slatkin M. A measure of population subdivision based on microsatellite allele frequencies. Genetics. 1995;139:457–462. pmid:7705646
- 27.
Nei M. Molecular evolutionary genetics. Columbia University Press, 1987, New York, NY, USA.
- 28. Nei M, Li WH. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proc Natl Acad Sct USA. 1979;76(10):5269–5273.
- 29. Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics. 1989;123:585–595. pmid:2513255
- 30. Fu YX. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics. 1997;147:915–925. pmid:9335623
- 31. Liao PC, Kuo DC, Lin CC, Ho KC, Lin TP, Hwang SY. Historical spatial range expansion and a very recent bottleneck of Cinnamomum kanehirae Hay. (Lauraceae) in Taiwan inferred from nuclear genes. BMC Evol Biol. 2010;10:124. pmid:20433752
- 32. Brower AVZ. Rapid morphological radiation and convergence among races of the butterfly Heliconius erato inferred from patterns of mitochondrial DNA evolution. Proc Natl Acad Sci USA. 1994;91:6491–6495. pmid:8022810
- 33. Mantel N. The detection of disease clustering and a generalized regression approach. Cancer Res. 1967;27:209–220. pmid:6018555
- 34. Clement M, Posada D, Crandall K. TCS: a computer program to estimate gene genealogies. Mol Ecol. 2000;9:1657–1659. pmid:11050560
- 35. Luo A, Lan H, Ling C, Zhang A, Shi L, Ho YW, et al. A simulation study of sample size for DNA barcoding. Ecol Evol. 2015;5(24):5869–5879. pmid:26811761
- 36. Egeland T, Salas A. Estimating haplotype frequency and coverage of databases. PLoS ONE. 2008;3(12):e3988. pmid:19098988
- 37. Feng X, Huang L, Lin L, Yang M, Ma Y. Genetic diversity and population structure of the primary malaria vector Anopheles sinensis (Diptera: Culicidae) in China inferred by cox1 gene. Parasites & Vectors. 2017;10:75.
- 38. Cameron SL. Insect mitochondrial genomics: implications for evolution and phylogeny. Annu Rev Entomol. 2014;59:95–117. pmid:24160435
- 39. Hao YJ, Zou YL, Ding YR, Xu WY, Yan ZT, Li XD, et al. Complete mitochondrial genomes of Anopheles stephensi and An. dirus and comparative evolutionary mitochondriomics of 50 mosquitoes. Sci Rep. 2017;7:7666. pmid:28794438
- 40. Walton C, Handley JM, Tun-Lin W, Collins FH, Harbach RE, Baimai V, et al. Population structure and population history of Anopheles dirus mosquitoes in Southeast Asia. Mol Biol Evol. 2000;17(6):962–974. pmid:10833203
- 41. Scarpassa VM, Conn JE. Molecular differentiation in natural populations of Anopheles oswaldoi sensu lato (Diptera: Culicidae) from the Brazilian Amazon, using sequences of the COI gene from mitochondrial DNA. Genet Mol Res. 2006;5(3):493–502. pmid:17117365
- 42. Chen B, Pedro PM, Harbach RE, Somboon P, Walton C, et al. Mitochondrial DNA variation in the malaria vector Anopheles minimus across China, Thailand and Vietnam: evolutionary hypothesis, population structure and population history. Heredity. 2011;106:241–252. pmid:20517346
- 43. Simon C, Frati F, Beckenbach A, Crespi B, Liu H, Flook P. Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reactions primers. Ann Entomol Soc Am. 1994;87(6):651–701.
- 44. Spencer PS, Barral JM. Genetic code redundancy and its influence on the encoded polypeptides. Comput Struct Biotechnol J. 2012;1(1):e201204006.
- 45. Sarma DK, Prakash A, O’Loughlin SM, Bhattacharyya DR, Mohapatra PK, Bhattacharjee K, et al. Genetic population structure of the malaria vector Anopheles baimaii in north-east India using mitochondrial DNA. Malar J. 2012;11:76. pmid:22429500
- 46. Yang M, Ma Y, Wu J. Mitochondrial genetic differentiation across populations of the malaria vector Anopheles lesteri from China (Diptera: Culicidae). Malar J. 2011;10:216. pmid:21810272
- 47. Makhawi AM, Liu XB, Yang SR, Liu QY. Genetic variations of ND5 gene of mtDNA in populations of Anopheles sinensis (Diptera: Culicidae) malaria vector in China. Parasites & Vectors. 2013;6:290.
- 48. Ray N, Currat M, Excoffier L. Intra-deme molecular diversity in spatially expanding populations. Mol Biol Evol. 2003;20(1):76–86. pmid:12519909
- 49. Mohanty A, Swain S, Kar SK, Hazra RK. Analysis of the phylogenetic relationship of Anopheles species, subgenus Cellia (Diptera: Culicidae) and using it to define the relationship of morphologically similar species. Infect Genet Evol. 2009;9:1204–1224. pmid:19577013
- 50.
Wang G, Li C, Zheng W, Song F, Guo X, Wu Z, et al. An evaluation of the suitability of COI and COII gene variation for reconstructing the phylogeny of, and identifying cryptic species in, anopheline mosquitoes (Diptera Culicidae), Mitochondrial DNA Part A, 2016; https://doi.org/10.1080/24701394.2016.1186665.