Introduction

Citrus is one of the world’s most important economic crops and the most widely grown fruit tree in Africa. Nigeria is the largest citrus producer in Africa and the ninth in the world with a production of about 4,088,994 tons from approximately 839,628 ha of groves in 2017 (FAO 2019). Oranges account for most of the citrus grown in the country, but significant quantities of grapefruit, tangerine, lime and lemon are also cultivated1. During the past four decades, citrus acreage and production have dramatically increased in Nigeria due to the suitable ecological and climatic conditions for its production and the establishment of the National Horticultural Research Institute (NIHORT) in 1975 to promote the horticultural sector of the country, with citrus as a mandate crop2.

This significant growth in citrus production in Nigeria stemmed mainly from an increase in acreage, as yields have remained mostly stagnant, not exceeding 4–5 tons/ha and far below the world average of 14–15 tons/ha3. Among the limiting factors of citrus productivity in Nigeria is the absence of large commercial production. Most of the citrus is grown by smallholder famers in mixed cropping systems with cocoa, kola, coffee or rubber, and with minimal or no investment in agrochemicals and other inputs1. Given the favorable tropical climate in the citrus production areas of Nigeria, arthropod pests and diseases may also constitute a major impediment to enhanced productivity. Among the arthropod pests affecting citrus, the sap sucking hemipteran pests are of great concern. Feeding damage by these insects due to uptake of plant sap causes distortion, wilting, premature leaf drop, reduction of fruit sizes, and occasional tree death4. Furthermore, some of these sap-sucking pests are documented vectors of pathogens of economically important citrus diseases such as Tristeza, Citrus Variegated Chlorosis, and Huanglongbing (HLB) or Citrus Greening, among others5,6.

Although citrus tristeza virus (CTV) has killed more than 50 million citrus trees on sour orange rootstock since the beginning of the epidemics in the 1930s in Argentina and Brazil5, HLB is undoubtedly now the most economically important citrus disease in the world6,7. HLB is a highly destructive and fast spreading disease that is affecting citrus industries in Africa, the Americas and Asia6,7,8. Three related phloem-inhabiting fastidious bacteria: Candidatus Liberibacter africanus (CLaf), Ca. Liberibacter americanus (CLam) and Ca. Liberibacter asiaticus (CLas) are consistently detected in HLB-affected citrus trees6. These bacteria are transmitted via clonal propagation or grafting of infected material and by two psyllid vectors. The African triozid (Trioza erytreae (Del Guercio) (Hemiptera: Triozidae)) is the primary vector of CLaf but it can also transmit CLas, while the Asian citrus psyllid (Diaphorina citri Kuwayama (Hemiptera: Liviidae)) is the vector for CLas and CLam9. To date, CLaf and CLam are restricted to the African and American continents, respectively, while CLas is widespread across countries in Asia, the Americas, and more recently in Africa10,11.

D. citri has invaded the Americas in the last two decades10,12,13,14,15,16. The rapid spread of CLas and its ensuing HLB is due primarily to the prolific and invasive nature of D. citri17. Diaphorina citri has also been reported more recently in Tanzania18,19 in East Africa, indicating its global spread. The presence of D. citri in East Africa poses significant risks of further spread across the African continent due to the increase in international and transcontinental trade. In addition to its invasive potentials, ACP can hitchhike budwood, rootstock seedling or nursery plants of rutaceous hosts19 transported across national and international borders. Post-introduction, the favorable climatic conditions prevailing across citrus-producing regions of Africa could favor its establishment and population growth.

Knowledge of the presence of D. citri as early as possible would facilitate the development and implementation of targeted pest and disease mitigation efforts to limit its spread and establishment. Despite being a major citrus-producing country in Africa, no previous studies have been conducted to determine the presence of D. citri in Nigeria. To fill this void, citrus commodity pest surveys were conducted in southwestern Nigeria to determine the presence or absence of D. citri and if present, to diagnose the sampled individuals for Liberibacter.

Results

Field detection and morphological identification

The 10 surveyed sites (groves = 4, backyards = 6) were all within altitudes ranging from 152 to 275 m above sea level. Using previously described features20,21,22, and based on comparisons with voucher specimens, the field-collected insect samples were morphologically identified as D. citri (Fig. 1). Voucher specimens of these samples have been deposited at the TAMUK-CC in Weslaco, Texas Entomology Laboratory, and at NIHORT, Ibadan, Nigeria. During the survey, D. citri adults were observed feeding on mature and/or young expanding citrus leaves from 80% of the sites (8 out of 10; grove = 2, backyard = 6), while nymphs were recovered only from one backyard site (10% of all sites). D. citri was not observed in two of the four surveyed groves. A total of 248 individuals (adult = 216, nymph = 32) were collected from the eight infested sites (Table 1). During the survey none of the inspected trees showed symptoms of infestation of African triozid, Trioza erytreae (Del Guercio) (Hemiptera: Triozidae) such as open galls on leaves associated with its feeding damage. In addition, no individual of the African triozid, was encountered. Considering the wide spatial distribution of the positive detection sites, and the observation of developing nymphs at one site, it is likely that D. citri is established and relatively widespread in Oyo state.

Figure 1
figure 1

Adult (A) and nymph (B) of the Asian citrus psyllid (Diaphorina citri Kuwayama) detected in different locations (Table 1) in Oyo state, Nigeria.

Full size image
Table 1 Field attributes of suspect Asian citrus psyllid (Diaphorina citri Kuwayama) individuals sampled from different locations across Oyo State, Nigeria.
Full size table

Molecular detection

Gene-specific DNA amplicons of the expected sizes (Table 2) were obtained from a subset of randomly selected 13 D. citri (9 adults and 4 nymphs) that were representative of the spatial diversity of the sampled insects. A total of 13 COI-specific (GenBank accession nos. MT040168 - MT040180), 4 argH-specific (MT036086 - MT036089), and 6 atpA-specific (MT040183 - MT040188) sequences were obtained. The BLASTN analysis of these sequences produced significant (≥99% nt identity; 100% query coverage; E-value 0.0) matches to corresponding gene-specific sequences of D. citri, Ca. Profftella armatura, and Ca. Carsonella ruddii available in GenBank from different countries. In pairwise comparisons, the D. citri-specific COI sequences derived in this study shared 99–100% nt identities among themselves and the same range of nt identities with corresponding global sequences of this psyllid species; the COI-specific D. citri sequences were significantly distinct (42–43% nt identity) from corresponding sequences of GenBank isolates of T. erytreae. The Ca. Carsonella ruddii argH-specific sequences derived in this study shared 99.8–100% nt identity with each other and 97–100% nt identity with corresponding global sequences of this psyllid P-endosymbiont. The Ca. Profftella armatura atpA-specific sequences derived in this study shared 99.5–100% nt identity with each other, and 99.5–99.8% nt identity with corresponding global sequences of this psyllid S-endosymbiont. These results provide definitive confirmation of the identity of the insect individuals as D. citri. They also showed that D. citri individuals from Nigeria carry the same primary and secondary bacterial endosymbionts species in their bacteriome as previously documented from other citrus-producing areas of the world9,23,24.

Table 2 Primers used for the amplification of gene segments of the Asian citrus psyllid (Diaphorina citri) and its endosymbionts Candidatus Carsonella ruddii and Ca. Profftella armatura.
Full size table

Phylogenetic analysis

The Tamura 3-parameter was determined as the model with the lowest BIC (Bayesian Information Criterion) scores and was therefore used in maximum likelihood (ML) phylogenetic analysis of each of the gene-specific sequences. As expected, the mtCOI sequences from Nigeria clustered within the D. citri clade, distinct from the T. erytreae clade on the psyllid ML tree (Fig. 2A). Further analyses revealed a clear segregation of the D. citri-specific mtCOI sequences into the previously defined Western and Eastern clades25 with strong (>60%) bootstrap support (Fig. 2B). Most of the adult and nymph mtCOI sequences from Nigeria (MT040168 - MT040172 and MT040174 - MT040180) (n = 12) segregated into the Western clade; only one sequence (MT040173 from Agbofieti, Ibadan) clustered into the Eastern clade (Fig. 2B). Notably, sequences belonging to both the Western and Eastern clades were present in this same grove location at Agbofieti, Ibadan (Fig. 2B). The argH-specific sequences of the P-endosymbiont Ca. Carsonella ruddii also segregated into the previously defined Western and Eastern clades25 with strong (>60%) bootstrap support (Fig. 2C). Like the ACP mtCOI sequences, most of the Nigerian argH sequences (MT036086, MT036087 and MT036089) (n = 3) clustered into the Western clade, while only one sequence (MT036088 from Agbofieti, Ibadan) clustered into the Eastern clade (Fig. 2C). The results also showed that the sequenced individuals from this location belonged to both the Western and Eastern clades (Fig. 2C). Unlike the mtCOI and argH sequences, the atpA sequences of the S-endosymbiont Ca. Profftella armatura segregated into three distinct clades with strong (>60%) bootstrap support (Fig. 2D). Interestingly, most of the sequences from Nigeria (MT040184 - MT040188) (n = 5) formed a distinct clade, now referred to as ‘African’ clade, separate from the Western and Eastern clades (Fig. 2D). An additional sequence from Agbofieti, Ibadan (MT040183) clustered into the Eastern clade apart from other sequenced individuals from this same location (Fig. 2D). These results showed that whereas gene-specific sequences derived from field-collected ACP individuals and their bacterial endosymbionts in Nigeria are genetically uniform, samples from one specific location (Agbofieti, Ibadan) represented distinct populations, a likely consequence of multiple introductions of the ACP into this location. None of the ACP samples tested positive for CLaf, CLam and CLas by qPCR.

Figure 2
figure 2

Maximum Likelihood (ML) phylogenetic trees depicting the evolutionary relationships between adults and nymphs of the Asian citrus psyllid (Diaphorina citri Kuwayama), and their primary and secondary endosymbionts, sampled from different locations (Table 1) in Oyo state, Nigeria and corresponding sequences of global populations of each organism. The ML trees were derived based on analyses of sequences specific to the mtCOI gene of D. citri (A and B); MT040168 - MT040182 derived in this study; others from GenBank), the argH gene of the primary endosymbiont Ca. Carsonella ruddii (C); MT036086 - MT036089 derived in this study; others from GenBank), and the atpA gene of the secondary endosymbiont Ca. Profftella armatura (D); MT040183 - MT040190 derived in this study; others from GenBank). The Tamura 3-parameter was determined as the model with the lowest BIC (Bayesian Information Criterion) scores and was therefore used in ML phylogenetic analysis for each of the gene-specific sequences (with 1,000 bootstrap replications). Branches with <60% bootstrap support were collapsed.

Full size image

Discussion

The results obtained in this study represent the first report of the invasive Asian citrus psyllid (D. citri) in Nigeria, to the best of our knowledge. Recently, the occurrence of D. citri was reported in East Africa from Tanzania18 and Kenya19. This has prompted ongoing intense surveillance for the pest especially in several citrus-producing eastern and southern African countries. Considering the risks associated with global movement of plant germplasm and the long history of citriculture in Nigeria, we initiated this study to determine the status of D. citri in Nigeria with the goal of detecting it early (if present) to inform concerted management efforts in Nigeria and across West Africa. Using a combination of morphological and molecular analysis, the occurrence of D. citri was confirmed from 8 spatially disparate locations in Oyo state, a major citrus-producing state in Nigeria that shares extensive land border with the Republic of Benin in West Africa.

The optimal temperature for D. citri development is 24–28 °C, but it has a larger temperature plasticity enabling it to survive temperature conditions ranging from 4 °C to 41 °C26. Using a temperature-based model of suitability, Taylor et al.27 reported that the entire continent of Africa has climate suitable for the establishment of D. citri and HLB. Their predictive niche map placed Nigeria and many West and Central African countries within the areas of greatest risk of HLB establishment should the disease enter these regions. Thus, the confirmation of D. citri in Nigeria in the present study highlights a grave threat, not only to Nigeria but also to other neighboring West African countries due to the invasive nature of ACP.

The mitochondrial COI (mtCOI) gene of D. citri is versatile for assessing the biodiversity within the insect’s population from different geographical regions and have been used extensively in population studies9,28,29,30. However, only few studies have assessed gene-specific sequences of Ca. Carsonella ruddii and Ca. Profftella armatura, both maternally inherited endosymbionts of D. citri (Wang et al.25 and references therein). Whereas the Ca. Carsonella ruddii like other primary endosymbionts may help provide essential amino acids that are present at low concentrations in the diets of D. citri31, Ca. Profftella armatura was reported as a defensive symbiont of the insect24. The important roles of these endosymbionts in the life of D. citri and the likelihood of their co-evolution with their host32 informed our decision to perform a comparative assessment of their encoded genes along with the host-specific gene to better decipher the evolutionary history of the sampled psyllid individuals from Nigeria. Our phylogenetic analysis revealed that the detected D. citri populations in Nigeria and their Ca. Carsonella ruddii endosymbionts mainly belonged to the Western clade that consists of psyllid populations that generally include individuals from Asia and the Americas25,33,34. It is therefore plausible that ACP may have been inadvertently introduced via infested plant materials from the Americas or Asia. However, unlike the congruent phylogenies of D. citri mtCOI and Ca. Carsonella ruddii argH (Fig. 2B,C), the Ca. Profftella armatura from Nigeria formed a distinct clade (Fig. 2D), indicating that they have diverged considerably from the taxa belonging to the Western and Eastern clade post-introduction. This may be an indication that D. citri may have been present long enough in Nigeria to allow for their local adaptation. Furthermore, the detection of a D. citri individual and corresponding endosymbionts belonging to the Eastern clade, a group that comprises populations restricted to Southeast Asia and South and East China, in Agbofieti, Oyo state and the occurrence of mixed variant populations in this location suggests that more than one introduction event of D. citri into Nigeria has occurred over time. This calls for more concerted efforts to strengthen quarantine measures to prevent further introductions of populations of this invasive pest into Nigeria and other West African countries.

Although citrus production in Nigeria has been expanding for the last few decades, there is a paucity of information on the occurrence and distribution of key citrus pests and diseases in the country. HLB has emerged as a global threat to the sustainability of citrus industries worldwide6 and the disease is a major concern for countries where smallholder farmers are the key players. While the HLB-associated bacteria are transmissible through grafting and vegetative propagation, psyllids play a key role in their long distance and within grove spread. Not only that D. citri can be moved over long distances via trade of citrus materials such as budwood, rootstock seedling or nursery plants19, it can naturally travel over long distances17,35. In many parts of the world, initial introduction and subsequent spread of the Asian form of HLB has mainly occurred through infective D. citri as was the case for Florida36, Texas37 and California10. The gravity of the possible consequence of HLB into Nigeria is underscored by the fact that the current subsistence nature of citriculture in Nigeria is antithetical to effective implementation of the recommended practices for HLB mitigation such as establishment of clean plant programs, area-wide intensive ACP management, establishment of larger contiguous blocks, etc.6,7. In addition, considering a recent report of CLaf from Nigeria45, the prospect of its mixed infections with CLas should be concerning6. Therefore, the findings reported here should trigger national and regional emergencies to determine the extent of D. citri occurrence in West Africa and to implement region-wide management efforts to control this destructive pest before it becomes endemic and prior to the incursion of CLas in the region.

Materials and Methods

Sample collection

Citrus pest surveys were conducted in groves and residential areas in October 2019 in Oyo state, southwestern Nigeria (Table 1) to determine the presence of D. citri. The survey sites were selected along major highways. A total of 10 sites (groves = 4; backyards = 6) were sampled. At each location, the citrus species was determined and a Garmin Montana 680t handheld global positioning device (Garmin, Olathe, KS, USA) was used to record the approximate sample coordinates. All the citrus trees present in backyards were inspected, while eight border trees were inspected per grove (two trees per grove perimeter). The canopy of each selected tree was visually inspected for the presence of D. citri life stages20. The young expanding flush shoots were preferentially inspected since they constitute the reproduction sites for D. citri17. The presence of feeding damage such as ‘epinasty’ or distortion of young and tender leaves due to sap uptake of D. citri nymphs and adults, and the presence of wax-like dropping were also used to guide inspection of flush shoots. Suspect D. citri adults were collected using an aspirator and nymphs were collected using a fine camel hairbrush into plastic vials containing 95% ethanol.

Identification of D. citri

Morphological identification

Initial visual identification of the insects was made based on their morphological characteristics20, following which a subset of samples was sent to Texas A&M University-Kingsville Citrus Center (TAMUK-CC), Weslaco, Texas, USA for further morphological and molecular analysis. A detailed morphological characterization of the suspect insects was conducted at the Entomology Laboratory of TAMUK-CC by comparing them with archival reference voucher specimens. Morphological identification of D. citri nymphs and adults were carried out following features described by Mead20, Yang21 and OEPP/EPPO22. Voucher specimens of identified D. citri nymphs and adults from Nigeria were deposited at TAMUK-CC and at NIHORT, Ibadan, Nigeria.

Nucleic acid isolation and PCR

Total nucleic acids (TNA) was extracted from individual ACP adults and nymphs according to the Dellaporta et al.38 extraction protocol. The TNA extracts were quantified, and their quality analyzed on a NanoDrop 2000 series spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, USA) and stored at −20 °C until use. A 2 µL aliquot of each sample was used as template in a 25 μL polymerase chain reaction (PCR) with the reagents and Rapid Protocol described for the PrimeSTAR GXLDNA Polymerase (Takara Bio USA, Inc., Mountain View, CA). The primer pair DCITRI COI-L and DCITRI COI-R were used to target an 821 bp fragment of the mtCOI coding region of each insect29 (Table 2). The DNA samples were also subjected to PCR using single copy housekeeping gene-specific primer pairs argH-F1/argH-R1 and atpA-F1/atpA-R1 targeting the argH of the psyllid primary endosymbiont (P-endosymbiont) Ca Carsonella ruddii and the atpA of the secondary endosymbiont (S-endosymbiont) Ca Profftella armatura, respectively (Table 2) to determine if the bacteriome of the sampled individuals from Nigeria contain these endosymbionts. DNA extracts from laboratory reared ACP individuals from TAMUK-CC were included as positive controls. The amplified products were ran on 1% agarose gels prestained with ethidium bromide along with the 100–2,000 bp Wide-Range DNA Ladder (Takara Bio USA, Inc.), followed by visualization under a UV-transilluminator.

Cloning and sequencing

The sample and target specific DNA bands of the correct sizes were excised and gel-eluted using the Zymoclean Gel DNA Recovery Kit (Zymo Research, Irvine, CA). The recovered DNA were ligated individually into the pJET1.2/blunt vector using the CloneJET PCR Cloning Kit (Thermo Fisher Scientific) as per the manufacturer’s recommendations. The ligation products were used to transform chemically competent DH5α Escherichia coli cells and two to three plasmids with PCR-verified correct size inserts per cloned DNA amplicon were isolated from recombinant E. coli cells using the GenElute Plasmid Miniprep Kit (Sigma-Aldrich, St. Louis, MO). Each plasmid sample was sequenced in both directions with the pJET1.2 F and pJET1.2 R primers by the Sanger method in a commercial facility (ELIM BIOPHARM, Hayward, CA, USA).

Bioinformatic analysis

The raw sequences were analyzed with VecScreen (https://www.ncbi.nlm.nih.gov/tools/vecscreen/) and trimmed to remove pJET1.2 contaminant. The CAP contig assembly program of the BioEdit software39 was used to derive a consensus sequence from each of the sample-specific forward and reverse sequences. Each of the consensus sequences was subjected to BLASTN analysis40 for species identification purposes. Gene-specific sequences representative of the taxon diversity were assembled from GenBank and the MUSCLE alignment program (http://www.ebi.ac.uk/Tools/msa/muscle/) was used to generate multiple sequence alignments for the sequences derived in this study and those obtained from GenBank. The gene-specific alignment files were used to determine the sequence identity matrices and for phylogenetic analysis with the maximum likelihood algorithm of the molecular evolutionary genetics analysis (MEGA) software version 7.041.

Tests for Ca. Liberibacter spp

The DNA extracts were assayed for CLaf, CLam and CLas using the Taqman Multiplex Real-Time PCR assays performed on ABI 7500 Fast Thermocycler (Thermo Fisher Scientific Inc., Waltham, MA) or a SmartCycler II (Cepheid, Sunnyvale, CA) and analyzed as previously described42,43,44. All reactions contained known positive and negative control DNA samples, and non-template water control. A sample was considered positive for a specific bacterium using a cycle threshold (Ct) of ≤37.