Telenomus remus, a Candidate Parasitoid for the Biological Control of Spodoptera frugiperda in Africa, is already Present on the Continent

The fall armyworm, Spodoptera frugiperda, a moth originating from tropical and subtropical America, has recently become a serious pest of cereals in sub-Saharan Africa. Biological control offers an economically and environmentally safer alternative to synthetic insecticides that are being used for the management of this pest. Consequently, various biological control options are being considered, including the introduction of Telenomus remus, the main egg parasitoid of S. frugiperda in the Americas, where it is already used in augmentative biological control programmes. During surveys in South, West, and East Africa, parasitized egg masses of S. frugiperda were collected, and the emerged parasitoids were identified through morphological observations and molecular analyses as T. remus. The presence of T. remus in Africa in at least five countries provides a great opportunity to develop augmentative biological control methods and register the parasitoid against S. frugiperda. Surveys should be carried out throughout Africa to assess the present distribution of T. remus on the continent, and the parasitoid could be re-distributed in the regions where it is absent, following national and international regulations. Classical biological control should focus on the importation of larval parasitoids from the Americas.


Introduction
The fall armyworm, Spodoptera frugiperda (JE Smith) (Lepidoptera: Noctuidae) is a highly destructive pest of cereals, and is a native of the tropical and sub-tropical regions of North, Central, In this paper, we report on the observation that T. remus is already present in Africa, speculate how it might have arrived on the continent, and assess the potential for its use in augmentative biological control in Africa. This is the first report of an egg parasitoid of S. frugiperda in the open field in Africa.

Field Sampling
Parasitized eggs of S. frugiperda were obtained from eight sites in five countries: Benin (2 sites), Côte d'Ivoire (1), Kenya (1), Niger (2), and South Africa (2), in 2017 and 2018 (Table 1). These collections were made as part of different, unrelated projects, during S. frugiperda population surveys or for sampling egg masses for laboratory rearing. Quantitative data on parasitism were not collected, as it was not the primary objective of these surveys. Sampling and rearing techniques varied between sampling sites, but in all cases egg masses of S. frugiperda were brought to the laboratory and held in small containers. Egg parasitoids that emerged from egg masses were removed from the containers and killed by immersion in 70-96% alcohol. Larvae from the non-parasitized eggs in the same egg mass were reared until adult emergence to confirm the identity of the hosts. Samples of the egg parasitoids were sent to the Centre for Agriculture and Bioscience International (CABI) and the Natural History Museum, London, UK (NHMUK) for morphological and molecular analyses. Voucher specimens of egg parasitoids from all localities are deposited permanently in NHMUK, the North-West University, Potchefstroom, South Africa; and CABI, Delémont, Switzerland.

Morphological Analysis
Identifying Telenomus species is particularly difficult among other minute parasitoids. Females tend to appear very similar between species, with only very few useful morphological characters. Male genitalia, however, provide fairly reliable characters for species identification. Specimen preparation and dissection and mounting of genitalia are relatively straightforward [27], and can be done successfully on specimens from which DNA has already been extracted [28]. Several male specimens (> 3) from each locality were dissected and examined.

Molecular Analyses
Thirteen specimens, obtained from all but one site (Maradi, Niger), were analysed (Table 1). To obtain barcodes (around 600 bp of the mitochondrial gene (mtDNA) cytochrome oxidase c subunit 1 (COI)) from the samples, we followed the protocols described in [29]. These included the isolation of DNA from fragments of the abdomen of each specimen, by adding 20 µL of microLYSIS ® -PLUS (MLP; Microzone Ltd., Haywards Heath, UK) to the material. The suspension was macerated with a sterile micropestle (VWR International Ltd., UK) to facilitate the disruption of the exoskeleton and tissues of the samples. DNA was then liberated into the MLP by placing the sample tubes in a thermal cycler and subjecting them to the heat profile recommended by the manufacturer for difficult samples. PCR reactions were carried out using a Hybaid PCR Express thermal cycler in heated-lid mode. Amplifications were carried out in 0.5 ml microcentrifuge tubes in 20 µL reactions containing: 1 µL MLP DNA extract; primers LCO1490 and HCO2198 (5'-GGTCAACAAATCATAAAGATATTGG-3' and 5'-TAAACTTCAGGGTGACCAAAAAATCA-3', respectively; [30]), each at 150 nM; and 10 µL of MegaMix-Royal (Microzone Ltd, Haywards Heath, UK) mastermix solution, containing optimized mixture of Taq polymerase in 2 × enhancing buffer (6 mM MgCl 2 ), with 400 µM dNTPs and blue MiZN loading dye. Reactions were made up to a final volume of 20 µL with sterile molecular grade H 2 O. PCR reactions were preincubated for 5 min at 95 • C followed by 39 cycles of 30s at 94 • C, 30s at 51 • C, and 75s at 72 • C. Samples were finally incubated for 10 min at 72 • C followed by chilling at 10 • C. A second round of amplification (i.e., "reamplification") was undertaken as follows: 1 µL of each of the above PCR products was used as a template. The reaction was carried out under the same conditions, with the exception of the number of cycles, which were reduced to 30. PCR products were visualized with gel electrophoresis, purified by microCLEAN (Microzone Ltd., Haywards Heath, UK), and resuspended in 15µL sterile molecular grade H 2 O. After sequencing reactions, excess unincorporated dye terminators were removed by means of DyeEx 2.0 (Qiagen, UK) gel filtration columns, according to the manufacturer's instructions. Eluted samples were resuspended in 16 µL highly deionized formamide (HiDi™; ThermoFisher Scientific, Gloucester, UK). Sanger sequencing was carried out on an ABI 3130 Genetic Analyzer (ThermoFisher Scientific, Gloucester, UK) in accordance with the manufacturer's instructions.
Sequences were aligned using the multiple sequence alignment plug-in CLUSTALW in MEGA7 [31]. Sequences obtained in this study were compared with authenticated sequences available from the Barcoding of Life Data system [32] and additional sequences from the GenBank ® data base [33] and trimmed to size. The evolutionary history was inferred by using the maximum likelihood method based on the Tamura-Nei model [34]. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the maximum composite likelihood (MCL) approach, and then selecting the topology with superior log likelihood value. Codon positions included were 1st+2nd+3rd+noncoding. All positions containing gaps and missing data were eliminated. Trees were drawn to scale, with branch lengths measured in the number of substitutions per site. The bootstrap consensus tree inferred from 1000 replicates is taken to represent the evolutionary history of the taxa analysed [35]. Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches [35]. Evolutionary analyses were conducted in MEGA7 [31].

Results
Morphological analysis of several male specimens from each of the eight collection localities suggested identical species with almost no variation. Specimens were compared with a series of T. remus from many localities in Asia and the Americas, and found to be conspecific. A single specimen of T. remus in the NHMUK collection from Kenya, reared from eggs of Spodoptera triturata (Walker), was identified as T. remus by the last author in 1988 [36]. Records of T. remus from Serbia (Europe) in the Hymenoptera online database [37] are actually from Venezuela. The specimen data read as follows: "Ex. huevo de Spodoptera spp.; Cria masiva Lab. Serbio; (Barquisimeto, Venezuela-Lara-m.) (30.VIII.1990)". "Serbio" is clearly the source of the error.
The thirteen barcode sequences derived from this study gave 100% match to each other and were identified by comparison with those available from the BOLD database. The sequences were deposited in GenBank with accession numbers MH681660-3 and MK533746-7; MK533750-4; MK533757-8. Over 99% of the pairs of bases were identical to a series of 31 specimens in BOLD identified as Telenomus remus. Matches varied from 99.05% to 99.63%, with 27 specimens from Bangladesh, and were 99.62%, 99.56%, 99.56%, and 99.34% with single specimens from Pakistan, Ecuador, USA (Florida), and Honduras, respectively, but only 96.3% with a T. remus specimen from India (Kerala). The samples from Honduras, Ecuador, and Florida are those described in detail in Hay-Roe et al. [38]. The most closely-related Telenomus species, for which barcodes are publicly available, is Telenomus goniopis, which overlaps with our specimens at 91.17-91.33% (Figure 1). There is a strong probability that the "T. remus" sequence from India is based on a misidentification.  [38]. The most closely-related Telenomus species, for which barcodes are publicly available, is Telenomus goniopis, which overlaps with our specimens at 91.17-91.33% (Figure 1). There is a strong probability that the "T. remus" sequence from India is based on a misidentification.  Table 1.  Table 1.

Presence of T. remus in Africa
This study shows that a species identified from both morphology and DNA sequences as T. remus has been attacking S. frugiperda eggs in at least five countries in West, East, and Southern Africa, and eggs of the lawn worm, S. triturata, in Kenya. The genetic distance between the specimens from Africa and the numerous Asian and American T. remus specimens available in the barcode datasets is sufficiently small (< 1%) to conclude that it is the same species as the one that was released previously in South Asia and the Americas. Even though differences between intra-and interspecific divergence strongly vary between taxonomic groups, and genetic distances have to be used with caution to define species, Virgilio et al. [39] observed that 95% of mean interspecific congeneric distances in insects were found in the interval 2.47-21.00%. In an unpublished thesis, Bowers [40] suggested that, in Telenomus spp., intra-specific barcode divergences were below 1.6% whereas interspecific divergences were greater than 5.6%.
It is not known when and how T. remus arrived in Africa, but it certainly arrived before S. frugiperda since a specimen was collected in Kenya in 1988. This species parasitizes several Spodoptera spp. and some other Noctuidae [41], and could have remained unnoticed on other hosts for a long time because the adults of Telenomus spp. are extremely small and difficult to distinguish and identify on a morphological basis [38].

Implications for the Biological Control of S. frugiperda
No matter the pathway of introduction, the presence of T. remus has important implications for the biological control of S. frugiperda in Africa. Firstly, it shows that the parasitoid complex of S. frugiperda in Africa should be studied in detail before the introduction of exotic parasitoids. Prior to this report, the only published record of field parasitism of S. frugiperda in Africa was by Sisay et al. [42] in East Africa, but they did not report T. remus or any other egg parasitoid, with the exception of the egg-larval parasitoid Chelonus curvimaculatus Cameron, found at two locations. There are efforts underway for importing parasitoids from the native range of S. frugiperda for releases in Africa, either for permanent establishment and control or for augmentative biological control. Telenomus remus is high on the list of priority species to introduce, because it is the main egg parasitoid of S. frugiperda in its native range and, especially, because it is already successfully used in augmentative biological control programs and therefore can easily be adapted to the African situation. Our finding that T. remus is already extant in Africa suggests that importation efforts should focus on larval or pupal parasitoids, since the other known egg parasitoids of S. frugiperda in America (mostly Trichogramma spp.) are too polyphagous to be considered for introduction into Africa and not as effective.
A specific research programme should be developed for T. remus in Africa. Surveys for egg parasitoids should be carried out to determine how widespread T. remus is in Africa, and parasitism rates should be properly assessed to evaluate its impact on S. frugiperda populations. A phylogeographic study using molecular tools, such as in [43], would help in clarifying the history of T. remus in Africa. These surveys should also determine which other African Lepidoptera species served as host for T. remus prior to the arrival of S. frugiperda. For instance, the single record of T. remus on S. triturata needs to be followed up closely. Knowing its host range in African agroecosystems may also support the development of measures to enhance the parasitoid in a conservation biological control programme. Surveys for egg parasitism in Africa may also reveal that other Telenomus spp. have adopted S. frugiperda, in which case their interactions with T. remus should be investigated. In Africa, at least 10 Telenomus species have been found on cereal stem borers and have been well studied [27,44]. In contrast, those attacking Spodoptera spp. are largely unknown.
If the distribution range of T. remus in Africa does not fully overlap that of S. frugiperda, the egg parasitoid could be redistributed in the regions where it is absent, following national and international regulations for introduction of natural enemies. Climate modelling, similar to [4] for S. frugiperda, should determine whether all areas in Africa are suitable for its establishment. Several studies have assessed its climatic requirements, e.g., [26,[45][46][47], which should allow for the development of such models. Methods to use T. remus in augmentative biological control programmes should be developed based upon those used in the Americas [10,20,23,25,26], and T. remus should be registered as biological control agent against S. frugiperda. However, local specificities need to be considered, e.g., the fact that, in the Americas, maize is often planted in large monocultures while most of the maize in Africa is cultivated by smallholder farmers in mixed systems. Challenges may also be encountered in the mass production of the parasitoid due to the inherent difficulty of rearing T. remus on its natural host and the need of a factitious host for mass production [23,26,48]. Since the T. remus populations established in Africa may have never encountered S. frugiperda until recently, and because it might have become adapted to completing development in yet unknown hosts, it may be less efficient on this host than populations from the Americas that have evolved with S. frugiperda during, at least, the past 40 years. Therefore, it would be interesting to compare T. remus populations of African and American origins regarding their performance on S. frugiperda in quarantine. If American populations appear significantly more efficient than the African populations, the introduction of American populations could be considered.

Taxonomy of Telenomus remus
Telenomus remus was described from Ulu Gombak, just NE of Kuala Lumpur, Malyasia [49]. In the original description, Nixon states: ''This species is probably Telenomus spodopterae Dodd, which was described from four females labelled "from eggs of a moth Spodoptera sp. on sugar beet, Krebet, Java, 23.7.1913". Nixon was reluctant to use Dodd's name, as he considered the original description inadequate and thought there might be an important difference in the fore wing proportions. At least two more Telenomus species appear morphologically indistinguishable from T. remus (including male genitalia): Telenomus nawai Ashmead was described from Gifu, Japan [50]; and T. soudanensis (Risbec) was described from West Africa (probably Mali) [51]. There is some likelihood that all of these named species are conspecific [44]. Further research to establish the identities and possible conspecificity of the three described species would focus on DNA sequencing of specimens collected recently from the three type localities. Two independent studies have found mating incompatibility between two apparent species: Raveendranath [52] found a population from Hawaii was reproductively incompatible with one from Barbados, and Arakaki et al. [53] found an isolated population of T. nawai in Japan that is thelytokous due to Wolbachia infection. Further research is needed to clarify the status of these species.

Conclusions
This study showed that T. remus, the main egg parasitoid of Spodoptera frugiperda in the Americas that is considered for introduction into Africa, has now been found in East, South, and West Africa. This finding has important implications for the development of augmentative biological control and IPM programmes against S. frugiperda in Africa.

Conflicts of Interest:
The authors declare no conflict of interest.