Trissolcus basalis (Wollaston) (Hymenoptera, Scelionidae) is an egg-parasitoid of the southern green stink bug, Nezara viridula (Linneaus) (Hemiptera, Pentatomidae). Many behaviors associated with female T. basalis host-finding and acceptance are mediated by chemosensory pathways, for which olfactory, gustatory and ionotropic receptors have been previously identified. Odorant binding proteins (OBPs) are small, globular proteins, one of the functions of which is the transport of odorant ligands through the aqueous lymph of chemosensory sensilla to these receptors. We identified 18 classical OBP sequences in the T. basalis genome and transcriptomes sharing an average 26.8% pairwise identity. Gene tree analyses suggest very limited lineage-specific expansion and identify potential orthologs among other scelionids and Hymenoptera. Transcriptome mapping and qPCR comparison of expression levels in antennae and bodies of both sexes determine that at least five TbOBPs are preferentially expressed in the female antennae. These are, therefore, prime candidates for further study to determine their role in detecting host-produced semiochemicals.

chemical ecology, chemosenses, expression profiles, OBP, olfaction, Platygastroidea, semiochemicals

The subfamily Telenominae (Hymenoptera, Scelionidae) encompasses nearly a thousand species of egg-parasitoids, many of which have proven to be valuable as biological control agents of their host insects (Laumann et al. 2008; Talamas et al. 2015; Sacaccini et al. 2020). Numerous studies have demonstrated that these minute wasps (generally 0.5–2.5 mm in total length) locate eggs to attack by orienting to semiochemicals produced by their adult hosts. These include sex pheromones, alarm pheromones, aggregation pheromones, and cuticular hydrocarbons (Isidoro et al. 2001; Yang et al. 2016). The host specificity of the wasps presumably is mediated by their ability to detect and respond appropriately to these chemicals. In gregarious species (the vast majority), telenomines mate on the host eggs from which the wasps emerge, suggesting that evolution of the chemosensory ability of the female wasp to find eggs may also be a driver of the speciation mechanisms that have resulted in their high species richness (Loch and Walter 2002).

Trissolcus basalis (Wollaston) is a parasitoid of the southern green stink bug, Nezara viridula (Linnaeus) (Hemiptera, Pentatomidae). In captivity, the wasps can successfully parasitize other hosts that are presented to them, but data from specimens collected in the field suggest that they specialize on parasitizing N. viridula (Johnson 1985). Behavioral experiments have demonstrated that females of T. basalis respond to the plant volatiles 3-methylbutanoic acid and p-benzoquinone (Slimani et al. 2016); to the N. viridula kairomones (E)-2-decenal, 4-oxo-(E)-hexenal, and the sesquiterpenes trans- and cis-(Z)-alpha-bisabolene epoxide; the stink bug cuticular hydrocarbon n-nonadecane; and to mucopolysaccharides on the surface of an egg mass (Bin et al. 1993; Mattiacci et al. 1993; Colazza et al. 2007; Laumann et al. 2009).

The chemosensory mechanisms of T. basalis are only beginning to be explored. Previously, we characterized the full repertoire of receptor proteins in this species based on genomic and transcriptomic data: there are 170 olfactory receptors (Ors), 1 copy of Orco, 34 gustatory receptors (Grs), and 23 ionotropic receptors (Irs) (Chen et al. 2021). Several of the Ors are preferentially expressed in the antennae of adult females, suggesting that at least some of these proteins may be tuned to semiochemicals important for host location in the environment. Narrow tuning of the Ors could contribute to the host specificity of this wasp (Andersson et al. 2015).

Before reaching the receptors in the dendritic membrane of the olfactory sensory neurons (OSNs), environmental odorants first interact with another set of proteins, the odorant binding proteins (OBPs). Recent review papers discuss the relevance of OBPs and their role in insect olfaction (Pelosi et al. 2018; Rihani et al. 2021). OBPs may have a generalized function to facilitate the movement of a broad range of hydrophobic odorants to the OSN dendrite through the aqueous sensillar lymph (Vogt et al. 1991; Pelosi et al. 2005; Sun et al. 2016). A number of cases have documented that a subset of OBPs, the pheromone-binding proteins, are themselves narrowly tuned to bind with a small number of compounds (Du and Prestwich 1995; Campanacci et al. 2001; Zhou 2010). OBPs may thus act as an initial filter to limit which odorants can reach and be detected by a specific neuron. Narrow tuning of the response of an OSN may be a function of specificity of the Ors, of the OBPs, or a combination of the two (Andersson et al. 2015).

The objectives of this work were to identify and characterize the OBPs of T. basalis, to compare expression levels between males and females, and to determine which, if any, are more highly expressed in the primary olfactory organs, the antennae. The ultimate goal is to integrate information on OBPs with that of the receptor proteins to better understand the physiological mechanisms by which these specialist parasitoids are able to find their hosts in nature.

The repertoire size of OBP-coding genes varies across characterized Hymenoptera genomes, from 14 genes in Linepithema humile to 90 in Nasonia vitripennis (Vieira et al. 2012; McKenzie et al. 2014). Among insects, the number of OBP-coding genes ranges widely, with over 100 OBPs reported from both of the mosquitoes Culex quinquefasciatus and Aedes aegypti (Manoharan et al. 2013; Rehani et al. 2021). The repertoire of 18 Classical OBPs we found in the genome of Trissolcus basalis, therefore, is on the low end of this spectrum. There are several different motifs among OBPs; beyond the Classical OBPs, non-classical motifs include dimer-OBPs with two sets of 6 conserved cysteines, C-plus with 8 conserved cysteines, C-minus with 4 conserved cysteines, and Atypical with 9–10 cysteines and a long terminus (Zhou 2010). Although it is common to find OBPs that fall within non-classical motifs, it is not unusual to find only sequences with the six conserved cysteine sites consistent with the Classical OBP motif (Gu et al. 2011; Lagarde et al. 2011; Zheng et al. 2016).

The small number of OBPs in Trissolcus basalis limits the possible number and size of lineage-specific expansions in this species. Among the 18 TbOBPs and based on high bootstrap values, we have confidence for the existence of only one pair and one trio of genes. Therefore, in contrast with a species such as Nasonia vitripennis, evolution in the number of OBPs in Trissolcus has been fairly conservative. This is also apparent in our gene tree analysis, which resulted in few lineage-specific expansions among T. basalis OBPs.

McKenzie et al. (2014) recognized four groups of OBPs from ants that they named the single-copy ortholog group, paralog group 1, paralog group 2, and the OBP59a group. They concluded that the two paralog groups were specific to aculeate Hymenoptera. A grouping of Trissolcus OBPs with one member of paralog group 2, AmOBP7, is suggested by the topology of the tree in Fig. 2. The bootstrap values for the nodes linking TbOBP14, TbOBP7, and TbOBP6 to that Apis protein have 0% bootstrap support, indicating that this clustering is spurious. The 4% bootstrap support for the node linking AmOBP7 and TbOBP13 is extremely low and may indicate that this association is also trivial. McKenzie et al. (2014) also reported that the Nasonia genome encodes a gene in the OBP59a group, and on the basis of their tree of hymenopteran OBPs in the supplemental information, we concluded that this was NvOBP64. A key finding in our analysis is that we found that TbOBP2 groups with NvOBP64 (97%), and the two are likely orthologs. Finally, each of the six lineages of OBPs in the single-copy ortholog group identified by McKenzie et al. (2014) were reported to have counterparts in Nasonia as well as in other groups outside of Hymenoptera. Therefore, they concluded that these genes probably date back at least to the origin of Holometabola. In our data, however, three of these lineages –OBP5, OBP10, OBP11 – have no clear ortholog in T. basalis. Of the three remaining, TbOBP1 clusters with AmOBP1 (43%); TbOBP4 with AmOBP6 (12%); and TbOBP12 with AmOBP9 and CcOBP11 (46%). The “single-copy orthologs” may be ancient as suggested by McKenzie et al. (2014), but our data suggest that not all of them have been retained in a recognizable form in the Scelionidae.

Our measures of relative expression levels in tissues and sexes differed in a few cases between transcriptome mapping and qPCR experiments. Only one transcriptome for each combination of sex and tissue was sequenced, and a mapping of transcriptome reads for each such combination is, therefore, only a single point estimate of the true expression level. In contrast, the qPCR experimental levels were calculated on the basis on five biological samples. We have greater confidence in the qPCR results, and some of the values derived from the mappings may be individual extremes. It is also possible that, despite our efforts to maintain constant rearing conditions, the conditions experienced by the wasps prior to collection may have differed in some manner that affected gene expression.

All OBPs found in the genome of T. basalis were recovered in at least one of the adult transcriptomes. This suggests that no OBPs are expressed solely in the immature stages of these parasitoids. Insofar as at least some these proteins are involved in chemosensation, this is not surprising. The larvae of T. basalis develop within the eggs of their hosts, a relatively homogeneous and isolated environment. We would anticipate, though, that in their more generalized function as encapsulins or transporters of hydrophobic ligands, OBPs would still play an important role in nutrition and development in all instars of the wasps (Ishida et al. 2013).

Our motivation in this work is to better understand the chemosensory role of OBPs. Therefore, our attention was focused on those proteins expressed at higher levels in the antennae, the structure that has been demonstrated to be necessary for host-finding and acceptance (Safavi 1968). Transcriptome mapping suggested four male-biased antennal genes: TbOBP1, TbOBP4, TbOBP5, and TbOBP7. However, qPCR measurements resulted in no evidence for higher expression of any OBP in the male antennae. TbOBP1 was, in fact, expressed at significantly higher levels in the female antennae, and TbOBP4 was expressed at relatively high levels (in comparison with the bodies) in the antennae of both sexes. TbOBP5 and TbOBP7 could not be measured by qPCR. The initial transcriptome mapping suggested eight genes preferentially expressed in the female antenna: TbOBP8, TbOBP10–13, and TbOBP15–17. Measurements using qPCR confirmed this expression pattern for TbOBP10–11, TbOBP13, and TbOBP16–17. TbOBP8, TbOBP12, and TbOBP15 could not be measured with qPCR. Genes TbOBP16 and TbOBP17 group together in our gene tree analysis but are found on different scaffolds, suggesting that they did not arise from a simple gene duplication. On scaffold_159, TbOBP16 is flanked on either side by TbOBP4 and TbOBP5, further isolating it from TbOBP17. In contrast, TbOBP8, TbOBP10 and TbOBP11 are syntenic on scaffold_353. TbOBP10 and TbOBP11 also group together on the gene tree and have similar expression profiles, possibly suggesting a common regulatory element.

Up-regulation of a subset of TbOBPs in the female antennae suggests to us that the proteins play a chemosensory role in mediating female-specific behaviors. These may include mate recognition, host habitat recognition, long-range host finding, short-range host acceptance, and recognition of marking pheromones applied after oviposition (Salerno et al. 2017; Slimani et al. 2017; Foti et al. 2019). With the data available, it is still unclear what role these proteins might play in T. basalis. We would predict that TbOBP1 has some general function in olfaction. We base this conjecture on the conservation of this gene across all Hymenoptera. Salerno et al. (2012) identified components of a female-produced sex pheromone in the species Trissolcus brochymenae (Ashmead) that stimulates intense antennation and mounting by males. Notably, we found no indication in the qPCR experiments of any TbOBP that is more highly expressed in the male antenna. We recognize that functionally important olfactory proteins may be expressed in low levels, and that high expression is not a requirement to be important and functional.

McKenzie et al. (2014) concluded that, in ants, the OBPs were “likely involved in conserved aspects of olfaction” and that the OBPs “may not rapidly evolve to recognize species-specific signals” (emphasis added). We need more information on the OBPs in other, related telenomines to be able to assess the birth and death rates of these genes and whether a subset of them have been similarly conserved in the evolution of the subfamily.

We thank Gerry Carter (The Ohio State University) for valuable discussions and advice concerning the statistical analysis of the qPCR experiments; and George Keeney and Jeni Ruisch (The Ohio State University) for husbandry of the wasp, stink bug, and mealworm colonies.