Origins of divergence – The causes of pre- and post-zygotic isolation barriers in Spodoptera frugiperda – ORIGINS

The methods used during the first half of the project are described below:
Population genomics (obj 1):
we have developped a new population genomics approach to overcome the limitation of typical genomics methods, which are based on the mapping of Illumina short reads against a reference genome assembly. Usually, researchers discard all the ambiguous mappings to minimize the false positives by applying the filtering according to variant calling scores, , which is inherently limited by mapping scores. This filtering excludes a significant proportion of genomic sequences for the inference. As genes under positive selection are most likely to diverge, the mapping against selectively targeted sequences can be more difficult and likely to be excluded. Researchers usually ignore this mapping issue by assuming that genomic loci with ambiguous or no mappings are non-variant positions and that genome-scan will identify genetic variants of interest from whole genomes. This new method involves for each sample to generate a phased genome assembly, split the assemblies according to the phased blocks, then map these blocks against the available chromosome-sized assembly and finally perform a variant calling using GATK without filtering. Positively selected loci will be then identified.
Test and genetic determinism of oviposition choice (obj 3)
To test the hypothesis, that differential female oviposition choices underlie the strain-specific distribution on host plants, we have conducted the following experiment at the Everglades Research and Education Center (University of Florida) in Belle Glade, Florida (Figure 7):
1. F2 backcross families are generated as follows: corn-strain females are mated to rice-strain males, resulting in CR hybrids (as these hybrids are more prone to mate as female adults). The CR hybrids are backcrossed to both the corn-strain and the rice-strain, so that half of the resulting backcross females inherited their Z-chromosome from a corn-strain father and half from a rice-strain father (to allow addressing differences caused by the Z-chromosome).
2. From the hybrid matings, parents and some hybrid offspring are frozen for subsequent genetic analysis (trio binning). These will serve as reference genomes with high similarity to the samples.
3. Mass matings of F2 backcross females with males from both strains will be set up for 2 consecutive nights.
4. Release females in natura in big cages overlapping with grass and corn plots after having marked the strains with different fluorescent dye. Observation at night of the oviposition site. Collection of females having oviposited on corn or on grasses separately. Trio binning to generate genome assemblies from the two parental strains and the hybrids and GWAS to detect genetic variants associated to plant choice.

Regarding objective 1, we were able to collect FAWs on two corn fields but not on grasses, to extract HMW DNA for PACBIO HIFI sequencing for 36 samples. A Bioinformatic pipeline to generate genome assembly from long reads has been set up and validated using a female of our lab Sf-C colony to generate a high quality chromosomal level genome assembly v8 including the two Z and W sex chromosomes, which will be used to detect genetic variants. Evaluation of methods to optimize genetic variants using simulated genome data has been performed and published (Lefouili and Nam 2022). Detection of signatures of selection from long read based individual genome assemblies in the group of samples collected on corn in Florida is on-going. The signatures of selection from mapping of short reads on reference genome in samples of FAW collected on corn and on grasses (preliminary data of the proposal) have been analyzed and described in a paper in revision in BMC Evol. Biol (Fiteni E. et al.,, in review) and highlight 3 genomic loci under selection.A second field mission was performed in August 2022 to collect FAW on grasses.
Regarding objective 2. A field mission involving S. Yainna (French side) has been organized to collect FAW on corn fields and rice field in various geographic locations in Sénégal (Col. Thierry Brevault, CIRAD). Sampling was effective in corn field but FAW was not found in rice fields (the fact that in the invasive area FAW does not attack rice has been confirmed by others as opposite to the native area). Illumina whole genome resequencing and variant calling were over. Now, loci under selective sweeps are being identified. In absence of FAW on rice, a “common garden experiment” aiming at testing adaptive evolution of invasive populations compared to native populations on 4 varieties of corn (Benin, Senegal, Porto Rico, Canada ) has been started.
-Regarding objective 3, before the field mission in Florida, backcross F2 families have been generated from the two lab colonies in Germany and sent alive with pure strains to University of Florida, site of Belle Glade, where adjacent plots of corn and bermudagrass had been planted by US collaborators (MPICE). The oviposition behavior of females of the two lab strains after mating and marked with fluorescent dies has been observed at night in large cages overlapping the two plants. Since oviposition behavior proved inconclusive, mating behavior was observed and contrasted the mating behavior observed in the same lab colony before and after under lab conditions. Additional lab observations mimicking field conditions will be conducted at MPI-CE as well as field observations by collaborators in Florida.
For triobinning, a male pupa and it’s parents have been sequenced and assembled according to plan. The analysis is in progress.
Regarding objective 4. Most experiments will be conducted under lab conditions, as field observations have proved inconclusive (see objective 3)

Prospect
Obj1
A second field mission was planned for August 2022 to collect FAW on grasses to enable detection of signature of selection at a high resolution using long reads based individual genome assembly and phased genetic variants in FAW groups from the two plants, corn and grasses. This mission just occurred at the moment when I am writing this 18 months report and was successful since 139 larvae could be collected on grasses. We plan to compare the effect of Knock-outs with CRISPR/Cas9 (MPI-CE) to Knock-downs with RNAi (DIGIMI) of a single candidate gene with a visual phenotype. This will enable us to evaluate the possibility to use RNAi for some candidate genes for more rapid results than CRISPR/Cas9 and intensify the collaboration between the partners.
Obj 2
The second part of the common garden experiment with native population is currently being performed in DGIMI lab with the same corn varieties on whole plants and similar conditions.
Obj 3
Additional lab observations mimicking field conditions will be conducted at MPI-CE as well as field observations by collaborators in Florida.
Obj 4
Most experiments will be conducted under lab conditions, as field observations have proved inconclusive (see objective 3). An observation chamber has been equipped with LED lighting to better mimic natural light conditions like dusk and dawn and allow for experiments with insect-plant-interactions. Results of these lab experiments might give rise to additional field experiments of the German partner at a later time point.

Publications
1. Lefouili and Nam, 2022, Scientific Reports, doi: 10.1038/s41598-022-15563-2
2. Durand et al, 2022, Incipient speciation between host plant strains in the fall armyworm, BMC Evol. Biol. doi.org/10.1186/s12862-022-02008-7
3. Fiteni et al, BMC Evol Biol (under revision, submission ID : 7208a84f-e220-455c-8ecf-173a47388180)
4. Haenniger et al, in prep
Communications
1. E.d’Alençon. Invited speaker at the 53rd Annual Meeting of the Society for Invertebrate Pathology (SIP), 28th June -2nd July 2021 “Know your enemy: Integrative study of plasticity, adaptive evolution and speciation in the Fall armyworm”
2. K. Nam, EKC, 21st-July-2022, The process of biological invasion in the fall armyworm,
3. E.d’Alençon, the 11th workshop on Molecular Biology and Genetics of the Lepidoptera, 21-27 August 2022, Kolympari, Crete. Centromeres, transposable elements, piRNAs and heterochromatin in Spodoptera frugiperda
4. S. Hanniger. the 11th workshop on Molecular Biology and Genetics of the Lepidoptera, 21-27 August 2022, Kolympari, Crete. The genetics of timing differentiation

Speciation is the evolutionary process by which new biological species arise. There has been increasing acceptance of sympatric speciation models, where reproductive isolation and incompatibility between populations evolve in the face of gene flow; yet documenting this process remains challenging since gene flow reduction most likely results from a combination of factors. The reproductive barriers can be initiated either by divergent selection (that is, ‘ecological’ or sexual selection that creates extrinsic reproductive isolation) or by the evolution — through genetic drift, as an indirect consequence of selection, or through genomic conflict — of genetic incompatibilities that cause intrinsic reproductive isolation.
The project aims at deciphering the factors at play in the divergence of the two sympatric strains of the fall armyworm (FAW, Spodoptera frugiperda). This noctuid moth is a significant agricultural pest which spread over the globe within the last three years. It is an ideal model organism to investigate different isolation barriers in incipient species. While morphologically indistinguishable, the two strains can be differentiated on the basis of mitochondrial DNA (mtDNA) haplotypes, and they exhibit a number of genetic and phenotypic differences that can function as pre-or post-zygotic isolation barriers. They are named Corn, “C” and Rice, “R” strains from their preferences for different host plant ranges.
The objectives of the project are:
1) To identify unambiguously and characterize the loci diverging between the strains
Using a new population genomics approach, all loci diverging between the strain will be identified and classified according to the hypothetical contribution to speciation between prezygotic isolation (timing, oviposition preference, pheromones, and female hybrid frigidity) and post-zygotic isolation (larval preference, larval performance, female hybrid fertility), and they will be further characterized functionally using CRISPR/CAS9.
2) To study the origin of post-zygotic reproductive isolation in invasive populations
Preliminary data of invasive populations in Africa suggests fitness costs for corn-collected larvae on rice and vice versa, illustrating FAWs adaptive potential in invaded areas. We will analyze the genomic differentiation of these invasive populations.
3) To investigate the causes of strain-specific host plant differentiation
The preference for different host-plant ranges of strains in the original distribution area likely results from different oviposition choice or from differences in egg quality. We will determine the genetic basis of these traits by GWAS in field experiments.
4) To determine the ultimate causes of the strain-specific timing differentiation
The strains differ by their mating time at night. We will test experimentally if these differences can be related to variation in biotic factors connected to different habitats, e.g parasitoids or predator occurrence or plant defense rhythms.
This project gives the unique opportunity to carry out such an integrative study based on the combination of various approaches, study of phenotypes in natura and forward genetics in Germany and reverse genetics and functional genomics in France. This integrative study, by allowing clarification of the evolutionary status of the native populations of FAW C & R strains, is a prerequisite to understand the success of the invasive populations in their new environment. The question of the success of this highly polyphagous pest species in new habitats across the globe is one of high socio-economical impact that the project team starts to study together.