Introduction
Insect pests represent a major threat for global food security (Oerke, 2006). Pests control largely relies on the use of pesticides which, on the other side, are increasingly reported to have harmful ecological effects (Forister et al., 2019). Future crop yield stability will thus depend on plant breeding efforts to improve crop resistance traits against pests (Crall et al., 2018). Hypersensitive response (HR) is one of the most studied traits associated with plant defence against pathogens and pests and it is often underlined by single resistance (R) genes (Wagner et al., 2021). So far only a handful of R genes have been identified against insect herbivores, mostly consisting of cell surface receptors (pattern recognition receptors, PRRs) or intracellular receptors (nucleotide-binding leucin rich-repeat, NLRs) (Kouretlis & van der Hoorn, 2018). Resistance traits against insects based on R genes are mostly limited to piercing-sucking insects such as gall midges (Bentur et al., 2016; Harris et al., 2012), or phloem-feeding insects such as aphids (Botha, Li, & Lapitan, 2005; Dogimont, Chovelon, Pauquet, Boualem, & Bendahmane, 2014; Klingler, Nair, Edwards, & Singh, 2009; Nicolis & Venter, 2018; Rossi et al., 1998; Sun et al., 2020), whiteflies (Nombela, Williamson, & Muñiz, 2003) and planthoppers (Tamura et al., 2014; Liu et al., 2015; Zhao et al., 2016). Remarkably, reports on resistance genes to chewing insects are even more scarce, with only a few studies showing that PRR surface receptors also mediate defenses against chewing caterpillars (Gilardoni, Hettenhausen, Baldwin, & Bonaventure, 2011; L. Hu et al., 2018; Steinbrenner et al., 2020). Indeed, immunity against feeding insect herbivores appears to be mainly controlled by polygenic quantitative traits (Kliebenstein 2017).
Given the lack of effective R genes against chewing insects, resistance mechanisms targeting insect eggs have been proposed as a complementary defence strategy (Fatouros, Cusumano, Danchin, & Colazza, 2016; Tamiru, Khan, & Bruce, 2015). Clearly, the recognition and killing of insect eggs is advantageous to plants as it prevents the destructive feeding by the hatching larvae (Hilker & Fatouros, 2015, 2016). The investigation of egg-killing traits thus represents an alternative and unexplored source of novel R genes to increase crop resistance to pests (Fatouros et al., 2016).
Cabbage white butterflies, such as the gregarious Pierisbrassicae and the solitary P. rapae (Lepidoptera: Pieridae), are pests of crucifer crops (Brassica spp.) and a serious agricultural challenge (Kumar, 2017; Ryan et al., 2019).Pieris eggs induce a HR-like cell death in their host plants of the Brassicaceae family resembling a HR induced by pathogens (Caarls et al., 2023; Griese et al., 2021; Shapiro & De Vay, 1987). Under field conditions, a severe cell death reduces egg survival up to more than 40% on black mustard B. nigra (Griese et al., 2021). Therefore, the egg-induced cell death represents a trait with a high potential as a novel plant defense against eggs, hence reducing the impact of successive larval stages. While it is known that plants respond toPieris eggs with a salicylic acid (SA)-dependent immune response (Bonnet et al., 2017; Bruessow, Gouhier-Darimont, Buchala, Metraux, & Reymond, 2010; Caarls et al., 2023; Little, Gouhier-Darimont, Bruessow, & Reymond, 2007), the genes involved in detection and activation of egg-induced HR remain unknown.
A few recent studies began to investigate the genetic basis ofPieris spp. egg-induced HR-like cell death using the model species A. thaliana (Groux et al., 2021) and the crop B. rapa (Bassetti et al., 2022), which both benefit from extensive resources for classical forward genetics. A genome-wide association study (GWAS) in A. thaliana identified two loci, an L-typelectin receptor-like kinase-I.1 (LecRK-I.1 ) and a putative Ca2+ channel glutamate receptor 2.7(GLR2.7 ) (Groux, 2019; Groux et al., 2021). Further, a QTL mapping in B. rapa identified three QTLs Pbc1-3 associated with cell death size (Bassetti et al., 2022). The QTLs included many genes involved in plant immunity, but they underlined large genomic regions yet to be fine-mapped. A partial overlap among the loci identified in both plant species was suggested, given thatBra LecRK-I.1 is included within Pbc3 (Bassetti et al., 2022). Overall, the study of plant-insect egg molecular interaction is a relatively recent field, and more research is clearly needed to understand to which extent the genetic regulation of Pierisegg-plant interaction is conserved between plant species.
Egg-induced HR-like cell death observed in A. thaliana andB. rapa manifests as a light necrosis that has no or little effect on egg survival (Groux, 2019; Groux et al., 2021). In contrast, egg deposition on B. nigra triggers a severe HR, spreading from the leaf abaxial up to the adaxial side, which correlates with a substantial killing of different Pieris eggs (Caarls et al., 2023; Fatouros et al., 2014; Griese et al., 2021; Griese et al., 2020). Previously, we found that the severity and occurrence of egg-induced cell death in B. nigra differs between plants of the same accession (Caarls et al., 2023) and between accessions (Caarls et al., 2023; Griese, Dicke, Hilker, & Fatouros, 2017; Pashalidou, Fatouros, Loon, Dicke, & Gols, 2015), suggesting the existence of natural variation for egg-induced HR. A treatment with a solution of compounds derived from the eggs, an egg wash, was shown to mimic egg-induced responses in plants (Caarls et al., 2023) and it provides suitable treatment to screen large genetic populations. In summary, B. nigra represents an ideal plant species to study the genetics of egg-induced HR because the phenotype is strong, stable, easy to score, varies between accessions, and has a proven egg-killing effect.
In this study, we investigated the inheritance and genetic basis ofP. brassicae egg-induced HR-like cell death in B. nigra . Given the lack of advanced genetic populations, we crossed plant material collected from the field. We found variation in the response between and within field-collected B. nigra accessions, and then used crosses between strong- and low-responding individual plants to study the inheritance of the trait. We found that Pierisbutterfly egg-induced HR B. nigra segregates as a Mendelian trait, and then performed genetic mapping through bulk-segregant analysis paired with whole genome sequencing (BSA-seq). Recombinant analysis was used to further fine-map the genetic region and identify a single dominant locus of ~50 kb which we namedPieris e gg-k illing (PEK). Within PEK, a gene cluster of TIR-NBS-LRRs, a type of NLR receptors was present, showed copy number variants (CNVs) between different B. nigragenomes.

Material and methods

Plant materials
The inheritance of the HR-like cell death was studied using B. nigra accessions collected from a local population in the floodplain of the Rhine River near Wageningen, The Netherlands (N51.96, E05.68). The accessions SF3-O1, SF19-O1, SF25-O1, SF29-O1, SF47-O1 and SF48-O1 originated from one multiplication by open pollination (“O1”) of accessions used in previous studies (Griese, Dicke, Hilker, & Fatouros, 2017). The accessions DG1, DG12 and DG29 originated from open pollinated wild plants collected in 2018. A single DG1 plant showing no HR in response to eggs was selfed to obtain accession DG1-S1. A single DG1-S1 plant was then crossed with a single SF48-O1 plant to obtain an F1 population (Fig. 2a). Single plants from the F1 that showed HR-like cell death were backcrossed to other DG1-S1 plants to obtain segregating backcross families (BC1). Selfing of individual plants was done to generate BC1-S1 and BC1-S2 populations.
Plants were grown in a greenhouse under standardized conditions (21° day / 18° night, RH 50 - 70%, LD 16:8 h). Seeds were vernalized at 4° C for two days to induce even germination and then were sown in small trays with sowing soil (Lentse potgrond, Lent, The Netherlands). Seedlings were transplanted one week after germination into 17 cm diameter pots with potting soil (Lentse potgrond, Lent, The Netherlands). Plants were grown for five weeks before treatment with P. brassicae egg wash.