http://orcid.org/0000-0003-0958-9432
Figures
Abstract
Citrus greening disease, or huanglongbing (HLB), is currently one of the most devastating diseases of citrus. The bacteria thought to be responsible for the disease, Candidatus Liberibacter asiaticus impact the majority of commercial citrus species worldwide. These bacteria are transmitted by the Asian citrus psyllid (ACP), Diaphorina citri Kuwayama, which is now found in most citrus growing regions. With no known cure, ACP-vectored HLB is responsible for significant economic losses to the global citrus industry. A better understanding of the global genetic diversity of D. citri would improve current and future pest management and mitigation programs. To assess the genetic diversity of D. citri in worldwide collections, a total of 1,108 sequences belonging to ACP gathered from 27 countries in the Americas, the Caribbean, Southeast and Southwest Asia were examined for the study. 883 D. citri came from 98 locations in 18 different countries, and were sequenced using a 678bp fragment of the mitochondrial cytochrome oxidase I (COI) gene. Additionally, 225 previously-reported D. citri COI sequences, were also included in our analysis. Analyses revealed 28 haplotypes and a low genetic diversity. This is in accordance with previous reports on the little diversity of D. citri in worldwide populations. Our analyses reveal population structure with 21 haplotypes showing geographic association, increasing the resolution for the source estimation of ACP. This study reveals the distribution of haplotypes observed in different geographic regions and likely geographic sources for D. citri introductions.
Citation: Fuentes A, Braswell WE, Ruiz-Arce R, Racelis A (2018) Genetic variation and population structure of Diaphorina citri using cytochrome oxidase I sequencing. PLoS ONE 13(6): e0198399. https://doi.org/10.1371/journal.pone.0198399
Editor: Tzen-Yuh Chiang, National Cheng Kung University, TAIWAN
Received: December 14, 2017; Accepted: May 20, 2018; Published: June 21, 2018
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: All sequencing files are available from the genbank database (accession number(s) (MH001373-MH001384).
Funding: This research was supported by a cooperative agreement, 12-8130-0162, between the United States Department of Agriculture - Animal Plant Health Inspection Service - Science and Technology (USDA APHIS S&T) and the University of Texas - Rio Grande Valley. Mention of trade names or commercial products in this publication is solely for the purpose of providing scientific information and does not imply recommendation or endorsement by the USDA; USDA is an equal opportunity provider and employer.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The Asian citrus psyllid (ACP), Diaphorina citri Kuwayama [Hemiptera: Liviidae], is a phloem-feeding insect that in addition to causing physical damages to the plant, is the most effective vector of Candidatus Liberibacter asiaticus, and Candidatus Liberibacter americanus, the causal agents responsible for Huanglongbing (HLB) disease [1]. Nymphs and adults of D. citri acquire the bacterial pathogen when feeding on infected plants. Once the bacteria are inside the vector, they replicate, invade the salivary glands, and thus enable Ca. Liberibacter to infect other citrus plants through psyllid feeding [2–4]. Citrus greening disease is considered the most serious disease of citrus in the world [5], as most commercial citrus species are adversely affected by this bacterium [6]. Affected trees develop a variety of symptoms including mottling of the leaves and chlorosis, root death [7, 8], and small, discolored fruit with a bitter taste [9]. At present, the primary methods to control HLB is destructive removal of infected trees, psyllid control and plantation of Ca. Liberibacter free trees [10, 11]. Therefore, as a key vector for Ca. Liberibacter, D. citri represents a dangerous threat to regions that are still free of this disease [12].
D. citri was first described in Taiwan in 1907, and it is thought to be native to Southwestern Asia [9, 13]. Specifically, it has been suggested that D. citri evolved in India, in association with a species of orange jessamine Murraya [14, 15]. It has since been documented in throughout north Asia, Southwestern Asia, and southeastern Asia, and in some parts of the Middle East, causing significant damages to the citrus growing regions in these countries [16, 17]. D. citri was first reported in South America in 1942, and now is common to citrus growing regions in Brazil, Argentina, Venezuela, Paraguay, and Uruguay [18]. It is also pervasive in Central America, reported in Mexico, Costa Rica, Belize and Honduras [17, 19], and most of the Caribbean islands including Bahamas, Cayman Islands, Cuba, Jamaica, Dominican Republic, Guadeloupe, Abaco Island, Grand Bahama Island, U.S. A. Virgin Islands, and Puerto Rico [17, 20].
ACP has also invaded the Pacific Islands of Hawaii, American Samoa, Northern Mariana Islands, and Guam and is present in the following states of the mainland United States: Alabama, Arizona, California, Florida, Georgia, Louisiana, Mississippi, South Carolina and Texas [16, 20, 21,22]. D. citri is being actively managed across the commercial citrus growing regions of the U. S. A., including Arizona, California, Florida, and Texas, as well as non-commercial regions. In Florida alone, citrus losses associated with D. citri and citrus greening has been estimated to exceed U. S. A. $3.6 billion in a 5-year period [23, 24], and its spread continues to threaten the entire citrus industry worldwide. Minimizing further infestations by D. citri is paramount to help reduce the spread of citrus greening disease.
Knowledge of the genetic variation and population structure of an invasive pest species, such as D. citri, can help on the identification of the geographical origin of the invader, and can potentially provide insights of invasion routes. This information can also be useful for biological control programs and other pest management programs [25]. Mitochondrial DNA analysis has been widely used for phylogenetic inferences and population genetic inferences [26, 27, 28]. Previously, the global genetic diversity of D. citri from 15 different countries was studied using mitochondrial cytochrome oxidase I (mtCOI) sequencing by Boykin et al. [27]. Their study detected 8 haplotypes divided in two major haplotype groups South West Asia (SWA), and South-East Asia (SEA). They proposed that two independent introductions occurred in the New World [27], as suggested by a previous study on the genetic diversity of D. citri [22]. However, due to the limited sampling in the Old World they could not resolve the geographic origin of D. citri. The authors suggested that further sampling, especially from the Middle East and Asian countries was needed to better define the genetic diversity of this insect.
Because of the limited knowledge on the genetic diversity of D. citri, identifying the source or pathway of the Asian citrus psyllid is difficult. Additional data from geographic regions where this pest occurs is needed. The objective of this study is to: (I) examine the genetic variation of D. citri using COI sequences in samples from different geographic locations and (II) to examine these collections for population structure in order to further populate the dataset used for the pathway analysis of this pest. This information would aid in decision-making concerning pest management programs.
Materials and methods
Insect collection
A total of 883 D. citri samples were collected from different citrus host-plants, and made available to us by collectors from 18 different countries between 2008 and 2012 (Table 1). Additionally, 225 D. citri COI sequences deposited in GenBank by Boykin et al. [27], Lashkari et al. [29], and Chaitanya et al. [30] were included (Table 2), for resulting in a total of 1,108 individuals included in this global phylogenetic analysis.
Genomic DNA isolation, Polymerase Chain Reaction (PCR) amplification, and sequencing
Total genomic DNA was isolated from each individual using the DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA). The primers (forward: DCITRI COI-L 5’- AGG AGG TGG AGA CCC AAT CT-3’) and (reverse: DCITRI COI-R 5’- TCA ATT GGG GGA GAG TTT TG-3’) designed by Boykin et al. [27] were used to amplify a partial fragment of COI from D. citri.
The Polymerase Chain Reaction (PCR) contained 16.77μl of DNAase free water, 2.50μl of 10X Taq Buffer (TaKaRa Bio Inc, Mountain View, CA), 2.60μl of dNTP mixture 2.5μM each (TaKaRa), 0.13μl TaKaRa Ex Taq polymerase (5U/μl), 1μl of each primer at 10μM and 1μl of DNA template for a total volume of 25μl. The amplification was conducted in Applied Biosystems (Foster City, CA) Gene Amp® PCR System 9700 thermal cycler and the following cycling parameters were used: 94°C for 2 minutes, followed by 35 cycles of 30 seconds at 94°C, 30 seconds at 53°C, of 1 min at 72°C extension and a final extension of 72°C for 10 minute, as described in Boykin et al. [27].
The PCR products were stained with Blue/Orange 6X loading Dye (Promega, Madison, WI) and visualized with ethidium bromide on 1.5% electrophoresis agarose gels at 90V for 90 minutes in 1X TAE buffer. Documentation of these gels was performed using UVP Gel-Doc-itTS2 imager (Upland, CA). Before sequencing, the PCR products were purified with Exo-SAP-ITTM (Affimetrix, Santa Clara, CA), using the company protocol. Sequencing of PCR products was performed using bidirectional sequencing and 3’ BigDye-labeled dideoxynucleotide triphosphates (v. 3.1 dye terminators, Applied Biosystems, Foster City, CA, USA), and run on an ABI 3730XL DNA Analyzer with the ABI Data Collection Program v. 2.0 at the Huck Institute’s Nucleic Acid facility at Pennsylvania State University. Sequences were edited using Sequencher® v. 5.0 (Gene Coders Corp. Ann Arbor, MI). The alignment was constructed in MEGA v. 6.0 [31], using Clustal W [32]. Sequences were trimmed of primers to 678bp.
Genetic diversity and population structure
Haplotypes (h) were identified using DnaSP v. 5.10.01 [33]. DNA Pairwise genetic distances were estimated in MEGA v.6. The number of haplotypes (h), haplotype diversity (Hd), nucleotide diversity (μ) the average number of pairwise differences (k) [34], and the Tajima’s D [35] and Fu and Li’s D and F [36] tests were determined with DnaSP v. 5.10.01 [33]. Tajima’s and Fu and Li’s tests were performed to test for neutral evolution and evaluate the potential for recent population expansion or contraction. General population genetics statistics were calculated to determine probable center of origins based on diversity. A haplotype network, which include haplotype frequencies, was constructed in PopArt [37] using the TCS statistical algorithm [38]. An analysis of molecular variance (AMOVA) was performed to determine the percentage of variation among geographic groups, among populations, and within groups, were performed in ARLEQUIN v. 3.5 [39].
Phylogenetic and haplotypes analyses
JModelTest 2 [40] was used to identify the best substitution model for the DNA sequence data. A Bayesian phylogenetic reconstruction was then conducted using Mr.Bayes v.3.2.3 [41] in Cipres [42] allowing transitions and transversions to have different rate. The analysis was performed using a Monte Carlo Markov Chain (MCMC) method. Four chains were run independently for 10,000,000 generations with sampling conducted every 1,000 generations. The COI sequence of the hackberry petiole gall psyllid (Pachypsylla venusta) was selected as an out-group for the analyses (accession number: NC_006157). Additionally, a phylogenetic analysis using maximum likelihood methods was performed in PhyML v. 3.0 [43], using the nearest-neighbor interchange tree rearrangement and 1,000 bootstrap replications. Output trees were visualized in Cipres with midpoint rooting.
Results
Genetic variation
The final DNA sequence alignment of mitochondrial COI from 1,108 individuals was 678bp and contained a total of 29 polymorphic sites, 12 parsimoniously informative sites, and no indels. There were 18 synonymous changes and 11 non-synonymous. Unique sequences were submitted to GenBank (accession numbers: MH001373-MH001384). This resulted in 28 haplotypes shown by samples gathered from among the 26 locations analyzed for the study (Table 3). Of these haplotypes, 16/28 (57.14%) were identified from single individuals (singletons). The genetic diversity for the entire sequence dataset was low (<1%), and the maximum number of substitutions was 5bp for the most diverse haplotype (Dcit-20-Texas). Nucleotide and haplotype diversity were 0.00293 and 0.734 respectively. Four haplotypes were observed among 95.84% of these individuals. These included Dcit-1 368/1108 (33.21%), Dcit-2 158/1108(14.27%), Dcit-3 160/1108 (14.45%) and Dcit-06 376/1108 (33.96%). The average number of haplotypes per collection was 2.19 and ranged from 1 to 10 haplotypes per site for the entire dataset. The majority (92.3%) of these haplotypes were private to a single geographic region: fourteen haplotypes were unique to South West Asia (SWA), five for North and Central America (NCA), four for South East Asia (SEA), and two for the Caribbean (CAR).
Phylogenetic analyses
The HKY85 + I + G substitution model was selected as the best model based on Akaike Information Criterion ranking. Modeltest revealed a proportion of invariable sites of 0.1860, a transition/transversion ratio of 1.6314, and gamma shape parameter of 0.6100. The Bayesian (Fig 1) and ML (not shown) phylogenetic analyses showed concordant topologies. The Bayesian analysis and the constructed phylogenetic network (Fig 2) support the presence of geographic structure for D. citri. This analysis revealed strong support for genetic clusters and groups. In the Bayesian analysis, two major genetic clusters were observed (A and B). Cluster A contains 14 haplotypes and shows strong support, and cluster B contains 14 haplotypes with moderate support (.67). Three genetic groups (B1, B2, and B3) were observed in cluster B. Each of these three groups showed strong branch support (Fig 1).
Two major clusters are observed: A and B, where B is further divided in groups: B1, B2 and B3.
Circles represent haplotypes and connecting lines represent patterns of relationship. Hash marks represent additional mutational steps. Colored circles represent haplotypes observed in the regions associated with their color (NCA–North and Central America, SAM–South America, CAR–The Caribbean, SWA–South West Asia, and SEA–South East Asia). Size of each colored circle is proportional to the observed frequency of the haplotype, and proportion of samples from each region is indicated by the proportion of color associated with that region. Black circles represent inferred haplotypes.
Population structure
Geographic association was observed for genetic clusters and groups structure (Fig 1). For example, in Pakistan 372/381 (97.64%) fall within cluster A, as well as 10/14 (71.43%), of the individuals from India, and 13/13 (100%) of the individuals from Colombia. This cluster included fourteen haplotypes: Dcit-06, Dcit-09 through Dcit12, Dcit-14 through Dcit-17, and Dcit-24 through Dcit-28. The haplotype network shows a star-like shape arrangement seen among haplotypes from cluster A. Dcit-06 is the most common central haplotype for this cluster with the remaining closely related haplotypes present in by no more than three individuals. The haplotypes are not more than three mutational steps from the central haplotype (Dcit-06).
Cluster B consisted of groups B1, B2 and B3. Group B1, includes five haplotypes: Dcit-01, Dcit-04, Dcit-05, Dcit-07 and Dcit-21. Dcit-01 is the most geographically distributed haplotype seen in the study. The remaining four haplotypes are not more than four mutational steps from the most frequent haplotype (Dcit-01), they are locally restricted to a single collection site each. This group included individuals from 14 of the 26 countries; U. S. A. 230/246 (93.49%), Mexico 40/40 (100.0%), Belize 9/9 (100%), Costa Rica 10/10 (100%), American Samoa 22/22 (100.0%), Saudi Arabia 8/8 (100.0%), Iran 10/10 (100.0%), Brazil 4/16 (25.00%), Puerto Rico 14/62 (22.58%), Pakistan 9/381 (2.36%), India 4/14 (28.57%), China 11/75 (14.67%), Indonesia 4/13 (30.77%), Reunion 1/7 (14.29%).
Group B2 includes five haplotypes (Dcit-02, Dcit-08, Dcit-18, Dcit-19, and Dcit-22). The most common and widespread haplotype is Dcit-02. The other four haplotypes are no more than two mutational steps away from Dcit-02, and are limited to a single collection site each. This group contain 7/7 (100.0%) of the individuals from Paraguay, 15/15 (100.0%) of the samples from Argentina, 6/6 (100.0%) of the samples from Uruguay, 61/75 (81.33% of the individuals from China, 4/4 (100.0%) of the specimens from Taiwan, 2/246 (0.81%) of the samples collected in the US, 10/10 (100.0%) of the samples from Vietnam, 9/13 (69.23%) of the specimens from Indonesia, 31/31 (100.0%) of the samples from Thailand, 3/3 (100.0%) of the individuals from Singapore, 5/5 (100.0%) of the individuals collected in Mauritius, and 6/7 (85.71%) of the specimens from Reunion.
Group B3 included haplotypes Dcit-03, Dcit-13, Dcit-20 and Dcit-23. The most common haplotype is Dcit-03. The three other haplotypes were not more than six mutational steps away and were limited to a single collection site each. This group contains 13/246 (5.28%) of the individuals from the US, 48/62 (77.42%), of the individuals from Puerto Rico, 8/11 (78.73%) of the samples collected in Guadeloupe, 72/72 (100.0%), of the individuals gathered in Barbados, and 19/19 (100.0%), of the individuals from Trinidad.
Using natural geographic barriers, previously posed hypotheses, and observed patterns of distribution, the data were delineated into five populations: North and Central America (NCA), South America (SAM), The Caribbean (CAR), South West Asia (SWA) and South East Asia (SEA). The AMOVA confirmed the significance of population structure for the five geographic populations (Table 4). The AMOVA rejected the null hypothesis that the five populations are homogeneous. The highest source of variation (75.92%) was observed within regions, according to the five populations we mention above, for those combinations we tested. Variation was also observed among populations (10.05%), and within populations (14.03%).
Values in bold are significant with p = 0.00000.
Regional population summary
South West Asia population.
In all, 413 individuals were successfully sequenced. A total of 16 haplotypes were observed in this region. We observed a higher nucleotide diversity (0.00447), and a lower haplotype diversity (0.218), as compared with the estimated overall nucleotide and haplotype diversity (Table 5). The most common haplotype recovered in this region is haplotype Dcit-6. This haplotype was observed in 364 (88.13%) of SWA individuals. This haplotype was observed in 361/374 (97%) individuals from Pakistan, 3/14 (21.42%) of the individuals from India, and 12/12 (100%) of the psyllids gathered from Colombia. We observed that most of the diversity for the entire dataset belonged within Pakistan and India. Collectively, the haplotype diversity for these two countries was estimated at Hd = 0.149.
Values in bold are statistically significant with p = < 0.02. The values on the columns represents Tajima’s D (D1), Fu and Li’s D (D2), Fu and Li’s F (F), sample size (n), average number of nucleotide differences (k), segregating sites (S), nucleotide diversity (π), number of haplotypes (h), and haplotype diversity (Hd).
Eight of the sixteen haplotypes were exclusive to Pakistan psyllids (n = 381). Dcit-11 was observed in three individuals from Pakistan and Dcit-16 was observed in two. The remaining five haplotypes (Dcit-9, -10, -14, -16 and -17) seen in samples from Pakistan were recovered from a single individual each. Among the 14 samples from India, six haplotypes were seen in seven psyllids and were found to be exclusive to that country. Haplotype Dcit-23 was observed in two individuals. The remaining five haplotypes (Dcit-24-28) were recovered from a single individual each. All the individuals from Saudi Arabia (n = 8) and Iran (n = 10) showed haplotype Dcit-1. Demographic analyses for this population revealed significant values for Tajima’s D (-2.27; P<0.02), Fu and Li’s D (-5.64: P<0.02), and Fu and Li’s F (-5.17: P<0.02) (Table 5). For SWA populations the Fst value was significant and relatively high (0.86623 P<0.05), suggesting that there is population structure within these populations.
South East Asia (SEA) population.
In all, 148 sequences were analyzed, and a total of six haplotypes were recovered for psyllids from this region. The haplotype (Hd) and nucleotide (μ) diversity for this region was of 0.354 and 0.00088, respectively. Haplotype Dcit-2 was observed in 118/148 (79.72%) of the SEA individuals, and it was the only haplotype observed in individuals from Taiwan (n = 4), Vietnam (n = 10), Singapore (n = 3) and Mauritius (n = 5). In China, five haplotypes were identified, three of which were exclusive to this country (Dcit-05, Dcit-08, and Dcit-22) and made up a small percent of the sample (4%, 1%, and 1%, respectively). The remaining haplotypes, Dcit-1 and Dcit-2, were observed in eleven individuals (14.67%) and fifty-nine (78.66%) individuals, respectively. Haplotype Dcit-18 was only observed in 9/31 (29.03%) individuals from Thailand. Demographic analyses in this region using haplotype (Hd) and nucleotide (μ) diversity revealed non-significant values for Tajima’s D (-0.94), Fu and Li’s D (0.08), and F (-0.31). The Fst value for this region was significant (P<0.05), and relatively high (0.80376), suggesting the existence of population structure.
North and Central America (NCA) population.
A total of 327 specimens were successfully sequenced for collections from this region, and seven haplotypes were observed. Six of the haplotypes were recovered from psyllids gathered in a single collection site each. The haplotype and nucleotide diversity of this population was of 0.734 and 0.00294, respectively. The majority of the samples from this region were observed in group B1 311/327 (95.1%). Haplotype Dcit-1 was the most common haplotype in this group; it was seen in 92.66% of the NCA individuals. This haplotype was seen in psyllids from the five geographical regions analyzed. In Edinburg, Texas, U. S. A., 13/119 (10.92%) individuals showed haplotype Dcit-3, consistent with samples gathered in the Caribbean. Haplotype Dcit-4 was only observed in 6 individuals and was unique to a specific location in Yucatan, Mexico. In Los Angeles, California, U. S. A., two individuals showed haplotype Dcit-19. Haplotype Dcit-7, Dcit-20 and Dcit-21 were observed in a single individual each, from psyllids gathered in St. Lucie, Florida, Edinburg, Texas and Los Mochis, Sinaloa, Mexico respectively. Demographic analyses using haplotype (Hd) and nucleotide (μ) diversity estimates resulted in non-significant results for Tajima’s D (-1.51), Fu and Li’s D (-1.37), and F (-1.71). The Fst value for this region was significant (P<0.03), but relatively low (0.10132), suggesting no population structure.
South America population (SAM).
Sequences from 56 individuals were included in the analyses. Only three haplotypes were recovered from individuals gathered in this geographic region. A total of 71.43% of the individuals carried haplotype Dcit-2, the most common haplotype found in SEA. This haplotype occurs in 44/56 of the analyzed individuals. These individuals were found in Argentina, Brazil, Paraguay, and Uruguay. Haplotype Dcit-2 is included in group B2. The second most common haplotype, Dcit-6, was found in all individuals from Colombia (12/12) and is common to SWA collections. Haplotype Dcit-6 in included in cluster A. Four individuals, all from a single location in Brazil, carried haplotype Dcit-1. Haplotype Dcit-1 is found in group B1. Demographic analyses in this population using haplotype (Hd) and nucleotide (μ) diversity revealed no significant values for Tajima’s D (1.60), Fu and Li’s D (0.88), and F (1.28). The Fst value for this region was significant (P<0.05) and relatively high (0.81727), suggesting the presence of population structure.
Caribbean population.
For the Caribbean region, sequences from a total of 164 individuals were analyzed, and three haplotypes were observed. Haplotype Dcit-3 was observed in 89.63% of the collected CAR samples. Additionally, this haplotype was recovered from 13 individuals from Texas. In Puerto Rico, fourteen individuals showed haplotype Dcit-1, which also occurs in NCA, SAM, SWA and SEA. Three individuals (27%) from Guadeloupe exhibited haplotype Dcit-13, which was unique to this region. Demographic analyses in this population using haplotype (Hd) and nucleotide (μ) diversity revealed no significant values for Tajima’s D (-1.12), Fu and Li’s D (0.89), and F (0.66). The Fst value for this region was significant (P<0.002), but relatively low, suggesting no population structure for this geographic region.
Discussion
Genetic diversity
A total of 1,108 DNA sequences from the COI mitochondrial region were analyzed, revealing the existence of 28 haplotypes and a haplotype and nucleotide diversity of 0.723 and 0.00294, respectively. Our results are in accordance with previous reports of the relatively low genetic diversity of D. citri in worldwide populations [27]. Because of this low genetic diversity, these results further suggest that D. citri is a single species and not a species complex. However, the population structure indicates barriers to gene flow. The numerous private haplotypes observed suggest differentiation and limited gene flow between the five different populations.
It is possible that this genetic isolation is caused by geographic barriers or biotic factors, such as infection with the endosymbiont Wolbachia. Recent studies indicated that 66% of all insect species may be infected with this bacterium [44], and Wolbachia has been found infecting D. citri. Infection with Wolbachia often causes reproductive abnormalities including cytoplasmic incompability (CI) [45]. It has been suggested that infection with different Wolbachia strains may lead to reproductive isolation between populations [46]. An association of COI sequences for D. citri and Wolbachia strains has been observed: U. S. A., Mexico, Belize and American Samoa D. citri samples were infected with Wolbachia strain ST-FL, as well as samples from Pakistan and Colombia [47]. These results suggest that NAM and SWA populations might be infected with the same Wolbachia strain. Similarly, D. citri samples from China, Singapore and Argentina carried Wolbachia strain ST-173, a different strain than the one from SWA and NAM. Another strain of Wolbachia was associated with samples from the Caribbean, Puerto Rico, Trinidad and Barbados [47]. These results further suggest possible isolation between populations infected with different Wolbachia strains.
The information provided in this study is also important for the development of effective biological control programs to control D. citri populations. The parasitoid Tamarixia radiata has been extensively used as a biological control agent for D. citri [48, 49]. T. radiata has been successfully introduced to several countries, including Reunion Island, Taiwan, Mauritius, Guadeloupe, Florida, the Philippines and Indonesia, with varied results in parasitism rates of D. citri [29, 50]. In Reunion, biological control with Pakistani T. radiata effectively reduced psyllid populations and helped to mitigate the impact of citrus greening [51]. However, when T. radiata (originating from Taiwan and South Vietnam) was released in Florida, the reported parasitism was lower than 20% [52, 53]. The use of genetic analysis that identify the different evolutionary lineages or strains of T. radiata and D. citri [54] may improve parasitism rates and effectiveness of these biocontrol programs. It is possible that T. radiata populations, as a biological control agent, are more effective for D. citri management when matched with pests that originate from the same geographic region. However, other ecological factors should be considered to understand variation in rates of parasitism [27].
Phylogenetic analyses and population structure
Based on the phylogenetic tree and haplotype network it was not possible to determine the ancestral haplotype for D. citri. Previous analyses in the geographic origin of D. citri had proposed Dcit-01 as the ancestral haplotype [27, 29], based on frequency and geographic distribution, and suggested a southwestern Asia origin. However, this assumption was based on limited sampling in this geographical region. Previously, only two haplotypes were reported for SWA: Dcit-01 and Dcit-06. In this analysis, additional samples from SWA were included, and a higher diversity was observed, as 13 haplotypes were recovered from this region. Even though Dcit-06 was the most frequent haplotype in SWA it has a limited geographic distribution as it was only observed in Pakistan, and in a few individuals from India and Colombia. The rest of the countries included in SWA (Saudi Arabia n = 8; Iran n = 10) expressed haplotype Dcit-01 only. More sampling from this region, especially from India, where the two haplotypes co-occur, is needed to confirm the geographic origin of D. citri.
SWA collections exhibited relatively high nucleotide diversity but low haplotype diversity. The haplotype network shows a star-shaped network for this region with a common central haplotype (Dcit-06), and a relatively high number of low-frequency haplotypes. These observations suggest that this is an expanding population, possibly caused by a recent introduction [55]. The diversity estimates generated for each region suggested a possible SEA origin. Relative to SWA (Hd = 0.218), SEA (Hd = 0.354) has high haplotype diversity, low nucleotide diversity, fewer segregation sites, and a high average number of nucleotide differences.
Consistent with results from previous analyses, Dcit-01 was the most geographically distributed haplotype in this study, and was present in all five geographical regions analyzed. However, it was more frequent in NAM populations (92.66%), when compared with the other geographical regions (SWA 7.51%; SAM 7.14%; CAR 8.55%; SEA 10.20%), suggesting a single introduction for NAM. This haplotype was the most frequent haplotype for all the collections from the New World (US, Mexico, Belize, Costa Rica), and it was also present and frequent for some countries in the Old World. In Saudi Arabia and Iran, Dcit-01 was the only haplotype found. It is possible that Dcit-01 originated in SWA, where it is relatively frequent. However, more sampling from this region is necessary to confirm this.
Haplotype Dcit-02 was the most common haplotype for SEA and SAM populations, suggesting that this Asian region is the geographic origin for South America psyllid populations. However, haplotype Dcit-01 was also present in Brazil, where it also co-occurred with Dcit-02, suggesting two independent introductions for this country. It is possible that Dcit-01 was introduced from SWA, or from China or Indonesia, where this haplotype co-occurs with Dcit-02. In addition, haplotype Dcit-06 was found in twelve individuals from Colombia, suggesting a possible third introduction for South America.
Haplotype Dcit-03 was found mostly in the CAR region with thirteen individuals showing this haplotype from Texas collections. Dcit-03 clustered within group B3 along with other haplotypes recovered from Texas (Dcit-20), Guadeloupe (Dcit-13) and India (Dcit-23). Since a closer relative has been found in India (Dcit-23), SWA could be a possible origin for this haplotype. Because this haplotype was found in some of the Texas samples, it is possible that a second introduction from the Caribbean to the U. S. A. occurred.
Even though relatively low genetic diversity was observed, the geographical distribution of the different haplotypes in our study support the existence of (1) five different populations within the different geographic regions: South West Asia (SWA), South East Asia (SEA), North and Central America (NCA), South America (SAM), and the Caribbean (CAR), and (2) four major haplotype groups (Cluster A, subclade group B1group B2 and group B3). A substantial portion of the haplotypes found in this study was unique for each region suggesting limited gene flow among the five geographic regions.
Acknowledgments
This research was supported by a cooperative agreement, 12-8130-0162, between the United States Department of Agriculture –Animal Plant Health Inspection Service–Science and Technology (USDA APHIS S&T) and the University of Texas–Rio Grande Valley. Mention of trade names or commercial products in this publication is solely for the purpose of providing scientific information and does not imply recommendation or endorsement by the USDA; USDA is an equal opportunity provider and employer. The authors would like to thank David Hall (USDA-ARS), Robert Shatters (USDA-ARS), Jesse De Leon (USDA-ARS), Matthew Purcell (USDA-ARS), Jeffry Makinson (CSIRO), Craig A. Webb (USDA-APHIS), Mark Schmaedick (American Samoa Community College), Emilio Arevalo Penaranda (Instituto Colombiano Agropecuario), Florinda Garcia Perez (Colegio de Posgraduados en Ciencias Agricolas, Texcoco), Isabel Lopez-Arroyo (Instituto Nacional de Investigaciones Forestales, Agricolas y Pecuarias–INIFAP), Manjunath Keremane (USDA-ARS), Aixa Ramirez (Universidad de Puerto Rico at Rio Piedras), Edda Martinez (Department of Agriculture of Puerto Rico), Siripor Donnua (Kasetsart University), Chandebham Shripat, Mamoudu Setamou (Texas A&M Kingsville–Citrus Center) and Marjorie Bestwick (USDA-APHIS), for assistance on D. citri collections.
References
- 1. Hall DG, McCollum G. Survival of Adult Asian Citrus Psyllid, Diaphorina citri (Hemiptera: Psyllidae), on Harvested Citrus Fruit and Leaves. Fla Entomol. 2011;94(4):1094–6.
- 2. Chu CC, Gill TA, Hoffmann M, Pelz-Stelinski KS. Inter-Population Variability of Endosymbiont Densities in the Asian Citrus Psyllid (Diaphorina citri Kuwayama). Microb Ecol. 2016;71(4):999–1007. pmid:26846216
- 3. Gottwald TR. Current Epidemiological Understanding of Citrus Huanglongbing. Annu Rev Phytopathol. 2010;48(1):119–39.
- 4. Wang NP, Setubal EA, Xu JC, Levy J, Zhang JG, Li Y, et al. The Candidatus Liberibacter-Host Interface: Insights Into Pathogenesis Mechanisms and Disease Control. Annu Rev Phytopathol. 2017;55(1):451–82.
- 5.
da Graca JV, Korsten L. Citrus huanglongbing review, present status and future strategies. In diseases of fruits and vegetables. Kluwer Academic Publiserhs, Dordrecht, Netherland. 2004;1:229–45.
- 6. McClean APD, Schwarz RE. Greening or blotchy-mottle disease of citrus. Phytophylactica. 1970;2(3):177–94.
- 7. Tsai JH, Liu YH. Diaphorina citri (Homoptera: Psyllidae) on Four Host Plants. J Econ Entomol. 2000;93:1721–5. pmid:11142304
- 8. Johnson EG, Wu J, Bright DB, Graham JH. Association of 'Candidatus Liberibacter asiaticus' root infection, but not phloem pluggin with root loss on huanglongbing-affected trees prior to appeareance of foliar symptoms. Plant Pathology. 2013;63(290–298).
- 9. Halbert SE, Manjunath KL. Asian Citrus Psyllids (Sternorrhyncha: Psyllidae) and Greening Disease of Citrus: A Literature Review and Assessment of Risk in Florida. Fla Entomol. 2004;87(3):330–53.
- 10.
Ayres AJ, Massari CA, Lopes SA, Bassanezi RB, Belasque J, Yamamoto P, et al. Current status of citrus HLB and efforts toward its eradication in Sao Paolo State, Brazil. Proceedings of 2nd International Citrus Canker and Huanglongbing Research Workshop. Florida Citrus Mutual, Orlando. 2005;H4.
- 11. López-Collado J, Isabel López-Arroyo J, Robles-García PL, Márquez-Santos M. Geographic Distribution of Habitat, Development, and Population Growth Rates of the Asian Citrus Psyllid, Diaphorina citri, in Mexico. Journal of Insect Science. 2013;13:114. pmid:24735280
- 12. Bove JM. Huanglongbing; a destructive, newly-emerging, century-old disease of citrus. J Plant Pathol. 2006;88:30.
- 13.
Mead F. The Asiatic citrus psyllid, Diaphorina citri Kuwayama (Homoptera: Psyllidae). Florida Department of Agriculture Conservation Service, Division of Plant Industry Entomology Circular 180 Florida. 1977.
- 14.
Beattie GAC, Holford P, Haigh AM, Broadbent P. On the origins of Citrus, huanglongbing, Diaphorinca citri and Trioza erytreae. Proceedings of the International Research Conference on Huanglongbing, December 2008 (ed by TR Gottwald & JH Graham). 2009(Plant Management Network, Orlando, FL, USA):23–56.
- 15. Hollis D. A new citrus-feeding psyllid from the Comoro Islands, with a review of the Diaphorina amoena species group (Homoptera). Syst Entomol. 1987;12:47–61.
- 16. Hall DG, Richardson M, Ammar E, Halbert SE. Asian citus psyllid, Diaphorina citri, vetor of citrus huanglongbing disease. Entomol Exp Appl. 2013;146:207–23.
- 17.
Hall DG. Biology, history and world status of Diaphorina citri. North American Plant Protection Organization, Proceedings of the Taller Internacional Sobre Huanglongbing y el Psillido Asiatico de los Citricos, del 7 al 9 de Mayo de 2008, Hermosillo, Sonora, p 11 Available at http://www.concitvercom/huanglongbingYPsilidoAsiatico/Memorıa-2%20Hallpdf. 2008.
- 18. Costa Lima AM. Insectos do Brazil: Homoptera. Escola Nacional de Agronomia Rio de Janeiro. 1942;3:141.
- 19. Halbert SE, Núñez CA. Distribution of the Asian Citrus Psyllid, Diaphorina Citri Kuwayama (Rhynchota: Psyllidae) in the Caribbean Basin. Fla Entomol. 2004;87(3):401–2.
- 20. Pluke RWH, Qureshi JA, Stansly PA. Citrus flushing patterns, Diaphorina citri (Hemiptera: Psyllidae) populations and parasitism by Tamarixia radiata (Hymenoptera: Eulophidae) in Puerto Rico. Fla Entomol. 2008;91:36–42.
- 21. French JV, Kahlke CJ, De Graca JV. First record of the Asian citrus psylla Diaphorina citri Kuwayama (Homoptera: Psyllidae), in Texas. Subtrop Plant Sci. 2001;53:14–5.
- 22. De Leon JS, Setamou M, Gastaminza G, Buenahora J, Caceres S, Yamamoto P, et al. Two separate introductions of Asian citrus psyllid populations found in the American continents. Ann Entomol Soc Am. 2011;104(6):1392–8.
- 23.
Hodges AW, Spreen TH. Economic impacts of citrus greening (HLB) in Florida, 2006/2007 to 2010/2011. Institute of Food and Agricultural Sciences, University of Florida. 2012:5.
- 24. Spreen TH, Balwin JP, Futch SH. An Economic Assessment of the Impact of Huanglongbing on Citrus Tree Plantings in Florida. Hortscience. 2014;49(8):1052–5.
- 25. Guillemaud T, Ciosi M, Lombaert E, Estoup A. Biological invasions in agricultural settings: insights from evolutionary biology and population genetics. C R Biol. 2011;334(3):237–46. pmid:21377619
- 26. Guidolin AS, Fresia P, Consoli FL. The genetic structure of an invasive pest, the Asian citrus psyllid Diaphorina citri (Hemiptera: Liviidae). PLoS One. 2014;9(12).
- 27. Boykin LM, De Barro P, Hall DG, Hunter WB, McKenzie CL, Powell CA, et al. Overview of worldwide diversity of Diaphorina citri Kuwayama mitochondrial cytochrome oxidase 1 haplotypes: two Old World lineages and a New World invasion. Bull Entomol Res. 2012;102(5):573–82. pmid:22717059
- 28.
Avise JC. Phylogeography: the history and formation of species. Harvard University Press, Cambridge 2000.
- 29. Lashkari M, Manzari S, Sahragard A, Malagnini V, Boykin LM, Hosseini R. Global genetic variation in the Asian citrus psyllid, Diaphorina citri (Hemiptera: Liviidae) and the endosymbiont Wolbachia: links between Iran and the USA detected. Pest Manag Sci. 2014;70(7):1033–40. pmid:24002991
- 30. Chaitanya BN, Asokan R, Rebijith KB, Krishna NK, Rao CN, Kumar PR. Molecular Identification and diversity of Asian Citrus Psyllids uisng mtCOI gene sequences. Ecosystems. 2016;22:51–7.
- 31. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Mol Biol Evol. 2013;30(12):2725–9. pmid:24132122
- 32. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucelic Acids Res 1994;22(22):4673–80.
- 33. Rozas J, Sanchez-DelBarrio JC, Messeguer X, Rozas R. DnaSP, DNA polymorphims analyses by the coalescent and other methods. Bioinformatics. 2003;19:2496–7. pmid:14668244
- 34. Tajima F. Evolutionary relationship of DNA sequences in finite populations. Genetics. 1983;105:437–60. pmid:6628982
- 35. Tajima F. Stattistical method for testing the neutral mutation hypothesis by DNA polimorphism. Genetics. 1989;123:585–95. pmid:2513255
- 36. Fu YX, Li WH. Statistical tests of neutrality of mutations. Genetics. 1993;133:693–709. pmid:8454210
- 37. Leigh JW, Bryant D. POPART: full-feature software for haplotype network construction. Methods in Ecol and Evol. 2015;6:1110–6.
- 38. Clement M, Posada D, Crandall KA. TCS: a computer program to estimate gene genealogies. Mol Ecol. 2000;9(10):1657. pmid:11050560
- 39. Excoffier L, Lischer HE. Arlequin suite ver 3.5: A new series of programs to perfomr population genetics analyses under Linux and Windows. Mol Ecol Resources. 2010;10:564–7.
- 40. Darriba D, Taboada GL, Doallo R, Posada D. jModelTest 2: more models, new heuristics and parallel computing. Nat Meth. 2012;9(8):772.
- 41. Ronquist F, Huesenbeck JP. MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 2003;19(1572–1574).
- 42.
Miller MA, Pfeiffer W, Schwartz T. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In: Proceedins of the Gateway Computing Environments Workshop (GCE), Loisiana, US. 2010.
- 43. Guindon S, Dufavard JF, Lefort V, Anisimova M, Hordijk, Gascuel O. New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst. Biol. 2010.
- 44. Hilgenboecker K, Hammerstein P, Schlattmann P, Telschow A. Werren JH. How many species are intected with Wolbachia?—a statistical analysis of current data. FEMS Microbiol Lett. 2008;281:215–20. pmid:18312577
- 45. Saha S, Hunter W, Reese J, Morgan JK, Marutani-Hert M, Huang H, et al. Survey of endosymbionts in the Diaphorina citri metagenome and assembly of a Wolbachia wDi draft genome. PLoS One. 2012;7:11.
- 46. Riegler MS, Stauffer C. Wolbachia infections and superinfections in cytoplasmically incompatible populations of the European cherry fruit fly Rhagoletis cerasi (Diptera, Tephritidae). Mol Ecol. 2002;11(11):2425–34. pmid:12406252
- 47.
Chu C, Hoffman M, Braswell WE, Pelz-Stelinski K. In Press. Genetic diversity and potential co-infection of Wolbachia among widespread Asian citrus psyllid (Diaphorina citri Kuwayama) populations. Insect Science.
- 48. Hoy MA. Classical biological control of Asian citrus psylla. Citrus Ind. 2001;81(48–50).
- 49.
Etienne J, Aubert B. Biological control of psyllid vectos of greening disease on Reunion Island. In Proceedings of the 8th Conference of the International Organization of Citrus Virologists Sidney, Australia. 1980:118–21.
- 50. Qureschi JA, Rogers ME, Hall DG, Stansly PA. Incidence of invasive Diaphorina (Hemiptera: Psyllidae) and its introduced parasitoid Tamarixia radiata (Hymenoptera: Eulophidae) in Florida citrus. J Econ Entomol. 2009;102:247–56. pmid:19253643
- 51. Chien CC, Chu Y. Biological control of citrus psyllid, Diaphorina citri in Taiwan. Int J Pest Manag. 1996;34:93–105.
- 52. Michaud JP. Natural mortality of Asian citrus psyllid (Homoptera: Psyllidae) in Central Florida. Biol Control. 2004;29:260–9.
- 53. De Leon JS, Setamou M. Molecular evidence suggests that populations of the Asian Citrus Psyllid parasitoid Tamarixia radiata from Texas, Florida, and Mexico, represent a single species. Ann Entomol Soc Am. 2010:10.
- 54. Barr NB, Hall DG, Weathersbee AA, Nguyen R, Stansly P. Qureshi A, et al. Comparison of laboratory colonies and field populations of Tamarixia radiata, an ectoparasitoid of the Asian Citrus Psyllid, using internal transcribed spacer and cytochrome oxidase subunit I DNA sequences. J Econ Entomol. 2009;102(6):2325–32. pmid:20069864
- 55. Aris-Brosou S, Excoffier L. (1996). The impact of population expansion and mutation rate heterogeneity on DNA sequence polymorphism. Mol Biol Evol. 1996;13(3):494–504. pmid:8742638