Patterns of spatial genetic structures in Aedes albopictus (Diptera: Culicidae) populations in China
The Asian tiger mosquito, Aedes albopictus, is one of the 100 worst invasive species in the world and the vector for several arboviruses including dengue, Zika and chikungunya viruses. Understanding the population spatial genetic structure, migration, and gene flow of vector species is critical to effectively preventing and controlling vector-borne diseases. Little is known about the population structure and genetic differentiation of native Ae. albopictus in China. The aim of this study was to examine the patterns of the spatial genetic structures of native Ae. albopictus populations, and their relationship to dengue incidence, on a large geographical scale.
During 2016–2018, adult female Ae. albopictus mosquitoes were collected by human landing catch (HLC) or human-bait sweep-net collections in 34 localities across China. Thirteen microsatellite markers were used to examine the patterns of genetic diversity, population structure, and gene flow among native Ae. albopictus populations. The correlation between population genetic indices and dengue incidence was also examined.
A total of 153 distinct alleles were identified at the 13 microsatellite loci in the tested populations. All loci were polymorphic, with the number of distinct alleles ranging from eight to sixteen. Genetic parameters such as PIC, heterozygosity, allelic richness and fixation index (FST) revealed highly polymorphic markers, high genetic diversity, and low population genetic differentiation. In addition, Bayesian analysis of population structure showed two distinct genetic groups in southern-western and eastern-central-northern China. The Mantel test indicated a positive correlation between genetic distance and geographical distance (R2 = 0.245, P = 0.01). STRUCTURE analysis, PCoA and GLS interpolation analysis indicated that Ae. albopictus populations in China were regionally clustered. Gene flow and relatedness estimates were generally high between populations. We observed no correlation between population genetic indices of microsatellite loci in Ae. albopictus populations and dengue incidence.
Strong gene flow probably assisted by human activities inhibited population differentiation and promoted genetic diversity among populations of Ae. albopictus. This may represent a potential risk of rapid spread of mosquito-borne diseases. The spatial genetic structure, coupled with the association between genetic indices and dengue incidence, may have important implications for understanding the epidemiology, prevention, and control of vector-borne diseases.
KeywordsAedes albopictus Microsatellite Genetic diversity Population structure Gene flow Dengue
analysis of molecular variance
polymorphic information content
Aedes (Stegomyia) albopictus (Skuse, 1894) originated in Southeast Asia and spread to several islands in the Pacific and the Indian Oceans during the 17th and 18th centuries . It was discovered in Albania in Europe in 1979 , the USA and Brazil in North and South America in 1986 [3, 4], Fiji in Oceania in 1990  and Nigeria in Africa in 1992 . Global warming, winter diapause, and human-aided transportation have contributed to the global invasion of Ae. albopictus [7, 8, 9].
As a vector of over 20 arboviruses  and one of the 100 worst invasive species in the world , Ae. albopictus poses serious public health concerns for arbovirus-related disease outbreaks. In China, dengue has been a threat for 40 years, since the first reported outbreak in 1978 . Local dengue transmission has been identified in Guangdong, Guangxi, Hainan, Yunnan, Fujian, Zhejiang and Henan provinces . Aedes albopictus and Ae. aegypti are two important vector species responsible for dengue transmission in China. Aedes albopictus is the most predominant species and has been found in nearly one third of China . This species has a wide range of distribution from north of 41°N latitude to the southern reaches of the country, while the distribution of Ae. aegypti is limited to small areas of southern China, including Hainan, Guangdong, Guangxi and Yunnan provinces [15, 16]. Li et al.  reported that Ae. aegypti was found only in two sites (Maoming, Guangdong and Sanya, Hainan) of the 26 mosquito surveillance sites in China. The relative proportion of Ae. albopictus to Ae. aegypti was > 20:1 in the two sites during 2005–2015, while the proportion of Ae. albopictus to Ae. aegypti was approximately 1:2 in 2014 in Jinghong city, Yunnan Province . Aedes albopictus has been reported to be the sole mosquito vector for dengue transmission in Guangzhou, with no presence of Ae. aegypti identified over the past three decades . The strong adaptability of Ae. albopictus has contributed to the reemergence and wide spread of dengue  and the outbreaks of mosquito-borne diseases have spread with the expansion of mosquito habitats . Vector surveillance, dispersal monitoring, and control play an important role in controlling outbreaks .
Molecular markers can provide abundant information that is critical to studying the sources and routes of vector invasions [11, 23]. As the second generation of DNA markers, microsatellites or simple sequence repeats (SSRs), consisting of short tandem repeats of 1–6 nucleotides, are widely distributed in the genomes of animals and plants . Due to the advantages of simple operation, easy detection, good reproducibility, high polymorphism, and the inheritance of Mendelian codominance , microsatellite markers have been utilized to evaluate genetic diversity and population structure, and identify invasions of Ae. albopictus [22, 26, 27]. Multiple microsatellite markers for Ae. albopictus have been isolated previously, and there are more than 50 microsatellites applied in population genetic studies of Ae. albopictus . Although mitochondrial DNA markers are rapidly evolving, non-recombining, maternally inherited, and exhibit high rates of mutation, mtDNA is not a strictly neutral marker; strong directional selection on mtDNA sequence has been reported in several insect species [28, 29, 30]. In order to obtain more accurate and detailed information on genetic structure, large-scale multi-site samplings should be conducted, and neutral microsatellite markers should be used to investigate Ae. albopictus population genetics.
The present study was designed to address the following questions: (i) What is the level of genetic diversity and population differentiation in and between Ae. albopictus populations in China?; (ii) What are the patterns of colonization and dispersal (or gene flow) of Ae. albopictus in China?; (iii) Is there any association between genetic diversity indices and local dengue incidence? To address these questions, extensive mosquito samplings and a set of thirteen microsatellite loci were used in the study. The data gained from this study could provide useful information for understanding the epidemiology, prevention, and control of vector-borne diseases.
Mosquito sampling and DNA extraction
Sampling information of Ae. albopictus collection in China
In each location, 8–12 households or collection points at 400–3000 m apart were selected randomly for adult mosquito collections to minimize sibling bias. Approximately 30 Ae. albopictus female mosquitoes from each location with 2–3 individuals per collection point were used for DNA extraction and genetic analysis. Local dengue surveillance data from the 19 provinces during 2011–2015 were obtained from the website (http://www.phsciencedata.cn/Share/en/index.jsp) of the Public Health Science Data Center managed by the Chinese Center for Disease Control and Prevention. The 19 provinces include those with frequent dengue fever outbreaks in the past 30 years, including Guangdong, Guangxi, Yunnan, Zhejiang and Hainan, as well as provinces with dengue outbreaks in recent years, including Hubei, Hunan, Guizhou and Jiangxi. Other provinces have imported cases, including Shanxi, Shandong and Liaoning. The collected specimens were identified morphologically to Ae. albopictus, and samples were stored in 70% ethanol. Total genomic DNA was individually extracted using the Insect DNA Kit (Omega Bio-tek, GA, USA), following the manufacturer’s protocol.
Cryptic species identification and Wolbachia infection
Aedes albopictus cryptic species identification was performed by DNA sequencing of the mitochondrial gene cytochrome c oxidase subunit 1 (cox1) for the samples collected in Guangxi Province, where the cryptic species have been detected previously . The Wolbachia infection status of individual mosquitoes was examined by PCR amplification of Wolbachia ribosomal DNA using a previously published method .
A set of thirteen microsatellite markers was used to examine the patterns of genetic diversity, population structure, and gene flow among the native mosquito populations. These markers, which were developed in previous studies, included 3 dinucleotide and 10 trinucleotide loci [27, 32]. The PCR reaction mixture consisted of 40 ng genomic DNA, 7.5 μl 2× PCR Master Mix (Promega, Madison, WI, USA), 0.05 μl (10 μM) M13 tagged forward primer, 0.2 μl (10 μM) reverse primer, and 0.2 μl M13 tagged fluorescent dye (FAM, HEX or TAMRA), in a final volume of 15 μl. The PCR cycling conditions were as follows: 95 °C for 5 min; 30 cycles of 95 °C for 30 s, 58 °C for 40 s, and 72 °C for 45 s; 8 cycles of 95 °C for 30 s, 53 °C for 40 s, and 72 °C for 45 s; and a final extension at 72 °C for 10 min. PCR products were sent to the Beijing Genomics Institute (BGI, Shenzhen, China) and processed in an automatic sequencer ABI 3730 (Applied Biosystems, Foster City, CA, USA). Allele sizes for each locus were read with GeneMarker software (version 2.6.3) .
Analysis of pairwise relatedness
Pairwise genetic relatedness between individuals was examined by LRM estimator  and QGM estimator  using GenAlex v.6.5  to detect kinship within and between Ae. albopictus populations. The coefficient of relationship, r ≥ 0.25 for LRM and r ≥ 0.5 for QGM was used to define a full sibling relationship (parents and offspring, or siblings that share the same parents), and a value of 0.125 < r < 0.25 for LRM and 0.25 < r < 0.5 for QGM indicates a half sibling (one shared parent). In order to reduce the siblings bias within population for analysis of genetic structure, we selected only one individual from each putative full-sibling group within any population.
Genetic diversity and differentiation
Genetic variation within each locality was estimated in terms of the average number of alleles (Na), effective number of alleles (Ne), inbreeding coefficient (FIS), fixation index (FST), observed heterozygosity (Ho) and expected heterozygosity (He) using GenAlEx v.6.5 . The number of migrants per generation, or gene flow, was calculated using the following formula: Nm = (1/FST − 1)/4 . Polymorphic information content (PIC) across all sixteen loci was assessed using Microsatellite Toolkit . The possible presence of null alleles was checked at a population level for each marker using Micro-Checker (version 2.2.3) . To determine whether the individuals in each study location were sufficient for the research, the mean number of alleles per locus (Na) for each population was calculated with a rarefaction method using Allelic Diversity Analyzer (version 1.0) . Deviation from Hardy–Weinberg equilibrium (HWE) and linkage disequilibrium (LD) were computed using GENEPOP version 4.7 [41, 42]. Analysis of molecular variance (AMOVA) was performed using Arlequin version 18.104.22.168 . To detect bottlenecks, the Wilcoxon test for heterozygosity excess was conducted with a two-phase mutation (TPM, 70% proportion of stepwise-mutation model in TPM, 30% variance for TPM, 1000 iterations) in BOTTLENECK v.1.2.02 [44, 45]. Genetic landscape shape (GLS) interpolation analysis was conducted using Alleles In Space (AIS) software based on the pairwise genetic and geographical distance matrices . Results of GLS interpolation analysis were imported into Surfer software to produce the contour map . A Mantel test on geographical and genetic distance (pairwise phiPT) was performed in GenAlEx v.6.5 using 9999 permutations. The correlation between geographical and genetic distance was plotted and the correlation coefficient (r) as well as R-squared were estimated using GenAlEx v.6.5.
Geographical population structure was evaluated using the Bayesian clustering method in STRUCTURE v.2.3 , which identifies the most probable number of genetic clusters (K) and assigns individuals to these clusters. We conducted different runs using different datasets for further clustering. For each dataset, the most likely number of clusters (K) was determined by conducting 5 independent runs for each K, from K = 1 to the maximum number of populations included in the analysis. Each run assumed an admixture model and independent allele frequencies using a ‘burn-in’ value of 50,000 iterations followed by 200,000 repetitions. The optimal number of clusters (K) was determined using the Delta K method of Evanno et al.  in the online version of Structure Harvester v.0.6.94 . Then we used a ‘burn-in’ of 100,000 and runtime of 2,000,000 generations per iteration (20 iterations) for the optimal K-values . To compile data from the 20 iterations for the independent values of K, we used the Greedy algorithm in CLUMPP v.1.1.2 with 1000 replicates . The results were plotted in DISTRUCT v.1.1 . Relationships among populations were assessed using principal coordinates analysis (PCoA) in GenAlEx v.6.5.
Correlation analysis between genetic indices and dengue incidence
Dengue incidence was calculated as the number of cases per 100,000,000 people. Due to the high variation in the number of cases, the data were rescaled with a log-transformation using the equation Ln (µ), where µ is the number of cases per 100,000,000 people. Pearson’s correlation coefficient was used to analyze the correlation between dengue incidence after transformation and genetic indices, including allelic richness, private allelic richness, effective alleles, Shannon’s information index, observed heterozygosity and inbreeding coefficient. The correlation analysis was performed using SPSS version 22. A significance level at P < 0.05 was set for all statistical tests, and the sequential Bonferroni correction  was used when significant correlations were detected between the paired data.
Detection of Ae. albopictus cryptic species and Wolbachia infection
No Ae. albopictus cryptic species was detected in the samples collected from Guangxi Province. The cox1 haplotype was identical with the sequences on GenBank (KY765450, KY765452 and KY765455). All the samples showed positive infection of Wolbachia with type A and/or type B. The majority of samples (> 93%) were infected with type A and type B together, only a few samples were infected with type A (5.1%) or type B (1.6%) (Additional file 1: Table S1).
Microsatellite analysis of genetic variability and diversity
Genetic relatedness between samples in Ae. albopictus population
Total pairwise comparisons
Genetic indices for genetic markers of Ae. albopictus from China
No. of alleles
Estimated null allele frequency
Genetic structure and differentiation
Analysis of molecular variance of populations from seven different clusters
Source of variation
Sum of squares
Percentage of variation
P < 0.0001
FCT = 0.013
Among populations within groups
P < 0.0001
FSC = 0.013
Among individuals within populations
P < 0.0001
FIS = 0.146
P < 0.0001
FIT = 0.168
Correlation between population genetic indices and dengue incidence
The correlation of dengue incidence with six genetic indices among Ae. albopictus populations in different provinces
Private allelic richness
Annual dengue casesa
Pearson’s correlation coefficientb
Few studies have reported the population structure and genetic diversity of Ae. albopictus in China, and they have focused on using mitochondrial DNA markers to analyze genetic diversity, regional structure, mainly in populations in southern China, or they have used a limited sampling size [31, 55]. For example, Guo et al.  sequenced the mitochondrial DNA cytochrome c oxidase subunit 1 (cox1) gene of 140 Ae. albopictus from 14 populations in southern China, and Zhang et al.  sequenced cox1 of 119 Ae. albopictus from 17 populations in China. The results of these studies showed low genetic diversity and shallow genetic differentiation between some of the study populations. In the present study we included many new study sites and populations, such as those in the northern (SY, SJZ, TJ), western (MS, TN), and central areas (WH, JZ), etc. The 34 collecting sites were widely distributed, providing comprehensive information on the population structure of Ae. albopictus throughout China. The correlation of average allele richness and sample size showed that when sample size < 15, allele richness increased rapidly as sample size increased, whereas it increased only slightly when sample size > 25. The number of samples in most populations was within the optimum range (25–30) . The mean number of alleles per locus and the allelic richness were higher in southern populations than in northern populations, and the diversity of alleles gradually decreased from south to north. The climate in northern temperate areas may affect the life traits of Ae. albopictus. The period in which the climate is suitable for mosquito survival is longer in southern subtropical areas than in northern areas. The relatively low temperatures and dry climate in the north may not be suitable for mosquito survival, reproduction, and dispersal, resulting in lower allele richness and population diversity in northern populations [57, 58].
A major drawback of using microsatellite markers in population genetic studies is the potential impact of null alleles, such as reducing population genetic diversity and increasing genetic differentiation among populations [59, 60]. Microsatellite null alleles are commonly more frequent in arthropods than in other species [61, 62, 63]; this may be a result of large effective population size, as species with large effective population size have relatively large proportions of individuals with null alleles . Many studies have determined that a null allele frequency < 0.2 has no significant effect on the accuracy of data analysis, and its effect on genetic diversity and genetic structure may be even less [64, 65]. In our study, null allele frequency was < 0.2 or absent, except at locus Aealbmic12. Expected heterozygosity is a common parameter for measuring the genetic polymorphism of a population . Takezaki & Nei  suggested that an expected microsatellite heterozygosity between 0.5–0.8 indicates that the population genetic polymorphism is high. In the present study, the expected heterozygosity in all populations was between 0.6–0.8, and most were higher than the observed heterozygosity, indicating a deficit of heterozygosity except in the SZ and SY populations . Insecticides have been used frequently in some areas, and fragmentation of habitats can intensify inbreeding within small populations, leading to a heterozygosity deficit. For bottleneck testing, we used the two-phase mutation (TPM) model, which is considered the best-fit for microsatellite data . Our bottleneck test results indicated that only the SJZ population may have experienced a temporary bottleneck; this may be related to the wide use of insecticides  or sharp climatic changes  in the local area causing a decline in local populations.
Our results showed that FST-values in all pairwise populations were not high, but most were significantly different, indicating genetic differentiation between populations. The low genetic distance and pairwise FST may be related to the fact that most of the study sites are central cities in their provinces and transportation between them is convenient. Jinghong, a busy tourist destination close to the Southeast Asian countries, connects to many cities via highways and airways but has no trains or high-speed rail, which explains the relatively high pairwise FST between Jinghong and other cities. Low FST-values with no significant difference may be attributed to the close distance, convenient transportation, high traffic, and similar environmental conditions between populations . China is home to an ancestral population of Ae. albopictus, and most populations in China have been stable for a long time . Nm > 1 indicates that a population is sufficient to prevent the occurrence of genetic differentiation . Our results showed strong gene flow between populations, which can reduce genetic differentiation. The results of genetic relatedness showed the individuals among different collection sites with long distance had different degrees of kinship, indicating potential human-aided dispersal of Ae. albopictus across the country. Extensive mosquito dispersal, probably aided by human transportation between these regions, also helps explain the low differentiation and the possible clustering between different regions.
There is some limitation for microsatellites compared to SNPs, the third generation DNA markers. Weaker separation between populations was usually obtained with microsatellites compared to SNPs. A total of 934 SNPs showed stronger power in distinguishing closely related individuals at a small spatial scale than 8 microsatellites in Ae. aegypti populations . A total of 2185 SNPs outperformed nine highly variable microsatellites in parentage and kinship assignment in An. darlingi populations, and the conventional approaches based on microsatellites may underestimate overall genetic distances in closely related vector populations . The pairwise FST or the degree of differentiation in the present study may be smaller for microsatellites than if SNPs have been used. Therefore, we might overestimate the gene flow between populations through the analysis of microsatellites. Mitochondrial DNA has been widely used for the study of molecular taxonomy, phylogenetic relationships and population genetics in mosquitoes [75, 76, 77]. However, microsatellites outperformed mtDNA in assessing spatial genetic structures. Only microsatellites not mtDNA showed small but positive and significant isolation-by-distance patterns in An. sinensis populations . No genetic structure was detected through mtDNA data while a high level of genetic structure was detected using the microsatellite markers inside polymorphic inversions in An. arabiensis populations . Guo et al.  discovered the cryptic species in Wuzhou using mitochondrial DNA cytochrome c oxidase subunit 1 (cox1) gene to analyze population genetics of Ae. albopictus in southern China. We also have collected Ae. albopictus population from a different village in Wuzhou, but we did not find any individual belonging to a cryptic species by cox1 sequencing. One of the possible reasons might be due to location-specific or seasonal abundance features of the cryptic species.
The spread of dengue fever has been affected not only by the climate (e.g. temperature, rainfall, relative humidity and sunshine) [80, 81], but also by the vector indices (e.g. container, house, Breteau, pupal or adult indices) and genetic factors of the populations [82, 83, 84]. Aedes aegypti population from French Polynesia was more susceptible to infection and had higher ability to transmit DENV-1 than Ae. aegypti population from New Caledonia in the same experimental environment . The genetic background of mosquito populations could influence the vector competence of mosquitoes, and the genetic indices can be utilized as a potential predictor of local dengue epidemiology . The six genetic indices are the main indices of population genetics, which could directly reflect the actual genetic variability and diversity of local mosquitoes. But they were not significantly correlated with dengue incidence during 2011–2015. One possible reason might be that dengue outbreaks are mainly affected by climate and environment rather than genetic background of Ae. albopictus population in China. The other one might be that dengue incidence is associated with other genetic indices or other genetic markers rather than the six indices from 13 microsatellites in this study. One genetic index, the inbreeding coefficient (FIS) might be expected to show significant positive correlation to dengue incidence, due to the vertical transmission of dengue virus to mosquito offspring [86, 87]. This natural transovarial transmission process has been demonstrated in field-collected Ae. aegypti mosquitoes from some different areas [88, 89, 90]. The rate of vertical transmission initially increased in the few generations (F1–F2) of Ae. aegypti mosquitoes, and in subsequent generations it was found to be steady until at least the 7th generation through the inbreeding of each generation . We inferred that the inbreeding within the mosquito population might lead to a substantial number of offspring carrying the dengue virus and an increased disease transmission within local areas. Inbreeding coefficient has been found to be positively associated with dengue incidence in Ae. aegypti populations . However, our study detected no significant correlation. One plausible explanation is that the amount of viral infection in the mosquitoes collected in our study is not enough to spread vertically to offspring. In addition, the endosymbiotic bacteria Wolbachia existing in Ae. albopictus, not in Ae. aegypti naturally, has a blocking effect on the vertical transmission of dengue virus [92, 93]. Although the specific mechanism of Wolbachia inhibiting dengue virus is unclear, the cytoplasmic incompatibility (CI) induced by Wolbachia can reduce the mosquito populations , and Wolbachia can induce density-dependent inhibition to dengue virus in mosquito cells . At present, mosquitoes transfected with Wolbachia have been applied in the control of mosquitoes and mosquito-borne diseases [96, 97]. The application would change the population structure and genetic indices of local mosquitoes, and it would require our monitoring. Further research is needed to ascertain on whether these genetic indices could be supplementary indices for the prediction of the local dengue epidemiology.
This study reported not only the spatial genetic structure of Ae. albopictus but also the correlation of genetic indices with dengue incidence. Our results may have implications for predicting future dengue outbreaks and understanding the epidemiology, prevention and control of vector-borne diseases. Strong gene flow probably assisted by human activities inhibited population differentiation and promoted genetic diversity among Ae. albopictus populations. This may represent a potential risk of rapid spread of mosquito-borne diseases. The data collected in this study, combined with other related factors such as climate, rainfall and human activities, will be valuable for vector control efforts as well as epidemiological prediction and modeling of the incidence and spread of vector-borne diseases.
We would like to thank Yuanhuan Wei, Ke Hu, Yulan Chen, Hao Zhang, Guangsheng Qiu, Weihao Zou, Yan Sun, Yijie Zhao, Denghua Wei, Weigui Ni, Jiayan Xie, Tianrenzheng Zhu and Yueyi Feng for their assistance in collecting mosquito samples from different provinces.
YW, XZ, DZ and GZ conceived and designed the experiments. YW, JW, ZS, YH, ZZ, PF and DY performed the experiments. YW and DZ analyzed the data. YW, XZ, DZ and GZ wrote and revised the manuscript. All authors read and approved the final manuscript.
This work was supported by the National Natural Science Foundation of China (No. 31630011) and the Science and Technology Planning Project of Guangzhou (No. 201804020084).
Ethics approval and consent to participate
No specific permits were required for the field studies. After explanation of the purposes and activities of the study, oral consent was obtained from the local participating residents prior to mosquito collection. No sites were protected by law and this study did not involve endangered or protected species.
Consent for publication
The authors declare that they have no competing interests.
- 26.Maynard AJ, Ambrose L, Cooper RD, Chow WK, Davis JB, Muzari MO, et al. Tiger on the prowl: invasion history and spatio-temporal genetic structure of the Asian tiger mosquito Aedes albopictus (Skuse, 1894) in the Indo-Pacific. PLoS Negl Trop Dis. 2017;11:e0005546.CrossRefPubMedPubMedCentralGoogle Scholar
- 77.Kamgang B, Ngoagouni C, Manirakiza A, Nakoune E, Paupy C, Kazanji M. Temporal patterns of abundance of Aedes aegypti and Aedes albopictus (Diptera: Culicidae) and mitochondrial DNA analysis of Ae. albopictus in the Central African Republic. PLoS Negl Trop Dis. 2013;7:e2590.CrossRefPubMedPubMedCentralGoogle Scholar
- 85.Carvajal TM, Ogishi K, Yaegeshi S, Hernandez LFT, Viacrusis KM, Ho HT, et al. Fine-scale population genetic structure of dengue mosquito vector, Aedes aegypti and its association to local dengue incidence. BioRxiv. 2019. https://doi.org/10.1101/561621. Accessed 28 Apr 2019.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.